TBPN Live - History’s Largest Oil Disruption, Oil & AI, Sundar's New Pay Deal | Alex Epstein, Dr. Alex Wissner-Gross, Charles Lamanna, Julien Bek, Eoghan McCabe, Michelle Volz

Episode Date: March 9, 2026

Sign up for TBPN’s daily newsletter at TBPN.com(02:43) - Ships in Gulf Declare Themselves Chinese to Dodge Attack (04:32) - Largest Oil Disruption in History Reactions (16:18) - Oil & AI (41:22) - Sundar's New Pay Deal (50:12) - Four Loko For Sale (01:02:01) - Dr. Alex Wissner-Gross, co-founder of Eon Systems, discusses the company's recent achievement in creating the first multi-behavior upload of a fruit fly brain, marking a significant advancement in whole-brain emulation. He explains how Eon integrated existing research, including the FlyWire connectome and NeuroMechFly simulation, to emulate the fruit fly's neural activity and motor functions, resulting in naturalistic behaviors like walking and grooming in a virtual environment. Dr. Alex emphasizes the potential of this work to pave the way for future emulations of more complex brains, such as those of mice and humans, aiming to enable human and non-human minds to operate in the cloud. (01:21:47) - Charles Lamanna, President of Business & Industry Copilot at Microsoft, leads the development of intelligent business applications and low-code platforms with generative AI. In his recent discussion, he introduced Copilot Cowork, an AI assistant that automates tasks across Microsoft 365 apps, integrating Anthropic's Claude model to enhance functionality. He also highlighted the upcoming Microsoft 365 E7 suite, which combines Copilot, Agent 365, and advanced security features to provide a comprehensive AI-powered productivity solution. (01:36:33) - 𝕏 Timeline Reactions (01:40:59) - Julien Bek, a partner at Sequoia Capital based in London, began his entrepreneurial journey in high school, building hardware companies before transitioning to software investments. In the conversation, he discusses the impact of AI on software businesses, emphasizing the shift from selling tools to delivering outcomes, and highlights the importance of combining AI's intelligence with human judgment to create defensible, service-oriented companies. He also shares insights on leveraging AI to enhance venture capital workflows, such as automating meeting analyses to improve decision-making and prioritization. (02:01:12) - Eoghan McCabe, co-founder and CEO of Intercom, discusses the company's recent $250 million funding round aimed at expanding their AI-driven customer service platform, FIN, which has achieved nearly $100 million in annual recurring revenue with 8,000 paying customers. He highlights the evolution of FIN from handling basic customer queries to addressing complex problems, emphasizing its ability to provide faster, more accurate, and cost-effective responses compared to human agents. McCabe outlines plans to extend FIN's capabilities across the entire customer lifecycle, transforming it into a comprehensive customer agent that enhances interactions from initial inquiries to post-purchase support. (02:14:34) - Alex Epstein, author of *Fossil Future*, discusses the critical importance of keeping the Strait of Hormuz open, as it facilitates the passage of approximately 20% of the world's oil supply. He emphasizes that any disruption could lead to significantly higher oil prices and global economic instability. Epstein also explores potential strategies to ensure the strait remains accessible, including military action, international alliances, and addressing threats like mines and drones. (02:44:45) - Michelle Volz, founder and managing partner of PAX, a $50 million early-stage venture fund, discusses her transition from Andreessen Horowitz's American Dynamism team to launching her own fund focused on foundational categories. She emphasizes the importance of founders being magnets for capital and talent, especially in sectors requiring significant early investment without immediate proof points. Volz also highlights the need for storytelling in hard tech industries, where traditional software metrics don't apply, and underscores the significance of progress in people, product, and traction to attract investors. (02:58:00) - 𝕏 Timeline Reactions TBPN.com is made possible by:Ramp - https://Ramp.comAppLovin - https://axon.aiCisco - https://www.cisco.comCognition - https://cognition.aiConsole - https://console.comCrowdStrike - https://crowdstrike.comElevenLabs - https://elevenlabs.ioFigma - https://figma.comFin - https://fin.aiGemini - https://gemini.google.comGraphite - https://graphite.comGusto - https://gusto.com/tbpnKalshi - https://kalshi.comLabelbox - https://labelbox.comLambda - https://lambda.aiLinear - https://linear.appMongoDB - https://mongodb.comNYSE - https://nyse.comOkta - https://www.okta.comPhantom - https://phantom.com/cashPlaid - https://plaid.comPublic - https://public.comRailway - https://railway.comRestream - https://restream.ioSentry - https://sentry.ioShopify - https://shopify.com/tbpnTurbopuffer - https://turbopuffer.comVanta - https://vanta.comVibe - https://vibe.coFollow TBPN: https://TBPN.comhttps://x.com/tbpnhttps://open.spotify.com/show/2L6WMqY3GUPCGBD0dX6p00?si=674252d53acf4231https://podcasts.apple.com/us/podcast/technology-brothers/id1772360235https://www.youtube.com/@TBPNLive

You're watching TVPN. Today is Monday, March 9th. We are live from the TVPAN Ultradome. Temple of Technology, Forteous of Finance, ramp.com, time is money, safe, budget, use corporate cards, bill pay accounting, and a whole lot more, all in one place. We have a fantastic show. We were able to get Alex Epstein to come down to the Ultradome in person

on short notice to get us up to speed on what's happening in the oil markets. Of course, oil is all, it's gushing all over the timeline, all over the financial times, all over. the Wall Street Journal as part of the... And Alex had a feeling that oil would be important in the future when he wrote the book, Fossil Future. Yes, which we have right here, which if you haven't given it a read, go pick a copy up, and stay tuned for him joining us at 110 today. Let's pull up the linear line up and tell you who we have coming on the show today.

Linear, of course, is the system for modern software development. 70% of enterprise work spaces on linear are using agents now. We have Dr. Alex Wistner Gross over on Eon Systems. He went very viral over the weekend for connecting a biologically derived fruit fly connectum to a Mujucco physics simulated body. And we're going to be talking all about simulating fruit flies and what will come after that. And we're also going over Microsoft co-pilot, co-waters. work, Julian. Which we're going to call Cocoa.

Coco, okay. Disney might like a word. They have IP in that space in particular, but we'll see. But do they have it in the enterprise? I don't think so. Julian's joining from Sequoia Capital. Own McCabe from Intercom. Our partner is joining because they just raised a massive round.

And Michelle Volz is launching Pax V.C. She left Andrews-N-Horwitz, I think about a year ago, and has been, is now launching a new early-stage fund. Anyway, let's go through the timeline and see how people are processing the news of the Strait of Hormuz and the oil price spikes. Fox News this morning, they were saying just basically their advice to the captain's to man up and just send it. Yeah, send it. Send it. That seems crazy. Effectively their advice. If you are on a shipping vessel, please stay safe. I'll read the actual quote. A guy Brian said, if you want to diminish the Iranian threat, if you want to make sure,

this ends with complete capitulation, show some guts and go through that straight. That seems very, very risky right now. Stay safe out there if you are on a shipping vessel, do whatever is responsible. There are some crazy twists happening. So apparently ships in the Gulf are declaring themselves as Chinese vessels to dodge attack. This is from the financial times. At least 10 vessels have changed transponder messages in an apparent attempt to avoid becoming target. A clutch of vessels trapped in the Gulf under enemy fire are adopting a tried and tested ruse to avoid attacks. They're changing flags, using transponders to declare themselves to be Chinese. There's always been a very odd, like, just like tug of war between how ships identify themselves.

Because often for tax reasons, they're bought in one country, operated by individuals from another company. but they fly a different flag to be able to go from one place to another, and it's all based on the port systems. I don't fully understand it, but it is very interesting. At least 10 ships over the past week have altered their destination signal to read Chinese owner, all Chinese crew or Chinese crew on board. About 1,000 ships are currently shut inside the Gulf and its immediate surroundings with the cumulative value of $25 billion.

And I don't know if you've seen some of the maps that show the straight, where it looks like nothing is actually moving through. I think in actuality, a number of the ships are actually turning off their transponders. So you can't see the movement because they're basically moving a little bit, going through the straight, and then turning back on again. Well, there are plenty of outlets that are covering the Iran conflict in detail. We, of course, will be focused on the business and technology impacts. I wrote about what oil prices mean for the AI buildout and data centers. It's just sort of interesting to dig into the deeper supply chain of artificial intelligence.

But there are some posts that we should go through around the oil story. So crude oil is five standard deviations above its 50-day moving average. Statistically speaking, this occurs every 9,500 years. So the last time would have been about 6,000 years before Moses parted the Red Sea. Imagine what that did to shipping in the area. Fanciful. What else is going on? COBSC, the COBSC letter has been posting a lot about oil prices.

U.S. oil prices are now expected to rise above $100 a barrel this month. That already happened. This was posted on March 6th. With markets pricing a 66% chance that it happens, the last time oil prices were above 100 was July of 2022. Let's pull up this clip from Landman. I haven't seen Landman. Have you watched it? Is it good?

Let's, uh... Well, you want oil to live above 60 but below 90. And don't get me wrong. We're still printing money at 90, but gas gets up over $350 a gallon. It starts to pinch. It hits 100. Every product in America has to readjust its price. $78 a barrel.

That's about perfect. It brings enough profit to keep exploring, but it don't sting as much at the pump. Unless, of course, you're in California. I mean, they tax the shit out of it out there. It could be $45 a barrel, and it's still $4 a pump. I don't know how those sandwiches do it out there. It's movie dad. 20, a barrel of oil was worthless.

This place became a ghost town. And nobody's immune. Kids have to quit college. Trucks get sold or repoed. We can pause it now. We got the gist of it. Yeah. Well, this is the largest.

How old were you when you discovered that gas is really expensive in California, not just because there's a lot of demand for it out here? I mean, I discovered it when I was filling up in Montana, and it was like $2 a gallon, and it's like $5 a gallon here. That was pretty wild. The first major gas price shock that I noticed was Hurricane Katrina in my life. I think it was in high school because I was too young. I was talking about just the fact realizing that it's so expensive because of the taxes that California puts on it.

Yeah, because it's nice to drive around here. Put the top down. The weather's good. So they're like, yeah, you're going to pay. You're going to pay more for the joys of driving an internal combustion engine car. Now, the price in California is aggressive. I don't know the structure of the prices, though. Is it percentage-based or fixed?

Because that affects how much the price will move based on oil price shocks. Because if the price per barrel doubles, but the tax is flat, you don't feel it as much here as you do other places. So, I don't know. Anyway. I'm pulling it up, but you can run through this. It's the largest supply shock by a factor of four. So the Hormuz blockade, which is current, 20 million barrels were lost in supply. The Iranian Revolution in 1978 was 5.6 million.

The Yom Kippur War embargoes in 73 was 4.4. The Iraq-Kuwait War was 4.3 in 1990. The Iran-Iraq War in 1980 was 4.0. And the Ukraine-Russia invasion in 2022, which is the last time that oil spiked over $100 a barrel, was 1 to 3. So an absolutely huge supply shock, and I'm sure it will have a lot of implications all over the economy. With triple digit prices, here's what's going to happen now, says policy tensor. Markets will scream when they open tomorrow. VIX will surge to levels beyond what we saw in April.

The sell-off will continue for some time as intermediaries shed risk. And the markets are red. They have been screaming today. the VIX futures curve already has inverted, bid up by dealers looking for insurance. This predicts a massive sell-off. The pressure on this captured White House now, the pressure on this captured White House now beings in earnest, that's kind of oddly written. Anyone can tell them that if this persists for very long, it will destroy the Trump presidency

and gut the GOP for a generation. The controlling factor here, as I have told you over and over again, is that the United States does not have the military means to reopen. Hormuz. There's no military solution in sight. This means that not only does Iran have the strategic upper hand now, it means that Iran enjoys the unambiguous strategic advantage. All they need to do is keep the thing closed until he capitulates. I put my neck out far to call this in advance. And someone told him yesterday I was in the minority, perhaps even a minority of one, no longer. I was correct. Just calling a shot. The blob heads and scribes were incorrect in their

assessment of the strategic situation and now markets will price to reflect reality. And we were, we briefly touched on Scott Sumner's blog post on Substack at the end of Thursday show, maybe Friday show. We didn't get a chance to read it. I actually read through it earlier today. It's pretty interesting. It basically makes the case, you know, it has a very controversial time to freak out and and the thought is that like he's doom posting. Maybe it's about AI. Maybe it's about this particular conflict. He is more just reflecting on this dynamic between when the market freaks out, it acts as a moderating effect to policy. And so he gives a number of examples around like tariffs. Or more specifically the

admin. Yeah. Yeah. Like the tariffs caused this massive circuit breaker, five percent sell off in the market. And then that was internalized and very quickly adjusted. And there were a whole bunch of different carve-outs to sort of like create soft landings. And so he's actually, and what he's getting at is that after the fact, a lot of people say, look, you didn't need to freak out because the taco happened. Trump always chickens out. The actual proposed policy effect didn't go into, didn't go into effect. And his point is that, well, it's precisely because people freaked out that it didn't go into effect. So freaking out is good in Scott Subner's mind, at least. Anyway, I love this. Yeah, we have a We have a solution. If you're feeling the pain at the pump, pivot, and get a horse. At oil

at 110, the urban horse is the only option. Pulling up to the gas station on a horse is truly elite. I do want to know what's the TCO on a horse with the food and the stables versus, you know, just keeping a, what is that, a Ford Taurus in your, I don't know, that's something else, you know, in your garage. The horse really mocks at the gas station in particular, right? Because it's just making everyone feel so stupid. Yeah, we got to go back. One horsepower is all you need. Nick Carter says,

seeing a lot of non-processed trusting panikins on the T.L. And then he followed up by saying, that's what I thought, because the oil prices spiked up, and then they fell down. And we did not get $120. a barrel, oil, we got exactly 100, which was still a bear. It's not like it's way, way, way too early to celebrate her.

Yeah, yeah, exactly. There's an interesting chart here, a really shocking chart from Oliver Grobe, unreal numbers about Germany's nuclear power generation. So if you scroll down, look at this curve. This is truly the bell curve meme or something like that. J.P. Morgan estimates that had Germany not phased out nuclear power, the country would have generated 50% less electricity from fossil fuels and 84% less electricity from natural gas in 2024. Electricity prices in Germany would have been around 25% lower and the country would have imported half as much electricity.

And just complete rise and fall of nuclear power generation in Germany. One of the craziest graphs. You don't see graphs like that very often in new technologies. Typically, you see S-curves or you see exponentials. No one considers the models get better and then they get dumber. That's certainly the funniest outcome. Yeah. Very, very rough. Goldman Sachs sent a note to investors saying if oil prices increase by $10 and remain elevated

for three months, U.S. year-over-year headline CPI inflation would likely rise from 2.4% in January to 3% in May. Those are small numbers, but we're looking at an oil pricing. increase of maybe $30, $40 over the baseline, potentially higher. We don't know where oil is going to land. And so you have this weird tug of war right now with the Fed where if oil prices go up, inflation goes up, the Fed has a mandate to curb inflation. That means higher interest rates. At a time when the labor market is shedding jobs, you would expect a Fed rate cut, or a lot of

People are optimistic about a rate cut. Trump certainly wants rate cuts. But if inflation is climbing, there's really no solution other than keeping rates high or even raising them further. So a real jam in terms of federal monetary policy, Fed monetary policy. Here's Art Cashin at the opening bell. This is a historical video.

When was this? This was a long time ago. Let's play this club. This may be it. First, let me start out. Moratorite Salu Thomas S.A. And you know that's the gladiator salute. We, who are about to die, salute you.

So it's going to be a tough morning. This may be it. We are about to die. Salute you. Insane, insane aura for, you know, CNBC really doesn't get enough credit for being so innovative in terms of broadcasting and entertainment. Really, some of the greatest clips.

I love one of the top comments. This is an old clip. Yeah, obviously. Obviously, it's 40P. It's probably from the 90s. So ZeroHedge says, you know, with oil at 111 total panic and Mimetic Cissifis shares eclipse, says half of this site for the last week has sounded like this. Let's see. Producers are telling me there is breaking news.

The Asian financial markets have just opened to a huge sell-off. And we're going to switch to that story right now. Good. I'm glad I'm here. Your thoughts, Tracy Jordan, on how this is going to impact Wall Street. Larry, I'm not an expert, but I do have a strong opinion. New York, as we know it, will no longer exist tomorrow. Producers are telling me there is...

Is that from 30 Rock? That's so good. That's so funny. Oil came way in from its overnight highs, says Joe Wisenthal. And the quote post is Chris Paul. It's a huge three to cut down the lead to 42. Absolutely. Absolutely brutal.

So let me tell you about gusto, the unified platform for payroll benefits and HR, built to evolve with modern, small, and medium-sized businesses. Stop making fun. We're somewhat of a smothern business. Small and modern. Small and modern. Let me also tell you about crowd strike. Your business is AI.

Their business is securing it. CrowdStrike secures AI and stops breaches. So we got to give you a shout out to George Kurtz for the Mercedes performance yesterday. Oh, yeah. One, two. One two. Really good. George Kurtz on that.

Would you look at that? Also. Take the stake in the team. Suddenly they're at the top. Yeah. He's on to something. Well, the, my essay this morning was titled, Why is no one talking about oil? Of course, everyone is talking about oil.

Oil discourse gushed onto the timeline this weekend. And crude prices spiked to nearly $120 a barrel as a broadening war in Iran threatens both transportation, routes, and production. The geopolitical and economic analyses are flowing. But what does this mean for AGI timelines? And a lot of people in the AI world are sort of tuning all of this out because they see recursive self-improvement, AGI, ASI, the buildout as more important. And I just wanted to sort of reality check the AI supply chain to understand how does oil

actually affect data center construction, AI production, token pricing? Like, is there any effect? My conclusion was that it's very moderate, but there are some interesting effects in the financial markets that are probably the bigger takeaway. But it's still interesting to hear about, like, yes, oil actually does, is used in the production of AI, at least a little bit. So power has been at the forefront of the AI pushback. Like everyone's worried about these local energy prices, these electricity prices increasing near the data centers. But, and it's been, it's become like a political issue. But pain at the pump might become a bigger story as gasoline prices spike. And that has been pain at the pump. Oh, it's been the most tangible sign of inflation and moves so

quickly, you know, one, one jitter in the economy. And it's a huge component. You know, people on the coast, people in tech don't have a good sense for this, right? If gas, gas for a lot of people, gas could quadruple and it wouldn't, they wouldn't really notice it. No, no. But if you actually look into the average American how gasoline fits into their budget, it's a meaningful component of their monthly budget, so they feel it super intensely. It's variable costs. There are so many different ways where a lot of Americans go on driving vacations.

That obviously is directly impacted by gas prices. And then also just psychologically, there's something about filling up at the gas tank where you see the number ticking up, and you're doing that on a every week basis or so, that it's just so visceral. It's this thing that you have to stop and then go experience and watch the money flow out of your account in real time.

It's very visual. It's very interactive. Yeah, I remember I remember I must have been probably 18 at this point where I would just go and I would, like, for a long time, I just put, I'd like, you know, prepay for a certain amount of gas. I got 20 bucks. I got 20 bucks. Let's see how much I'd get.

And I felt like really. Like I felt like the king of the castle once I just put my card down and let it run up. Yeah, yeah, yeah. Yeah. No, no, totally. So paint of the pump's been like a big, broad economic issue. It's so visceral. Like the actual phrase has a lot of power in politics and in like attack ads.

Like your gas prices are going up under this person. That's why you've got to vote for me. This is, this is common. But in AI circles, the discussion's been much more focused on RSI now. is the new acronym that everyone's focusing on? Not AGI. AGI's here. We know artificial general intelligence.

We pass the touring test. The models can do things generally, intelligently. But can they recursively self-improve? Are they RSI? Are we in RSI now? Is it a coming? Is this going to be a fast take-off? Is this going to be a slow take-off?

Well, something's taken off. Dylan Patel said being in SF is like being in Wuhan right before the pandemic. Something is happening. It's going to hit everywhere. but so few people know it. So he's sort of echoing that something big is happening mentality, this idea of recursive self-improvement, being AGI-pilled. A lot of smart people agree.

There's definitely something happening. There's still this question of like, what is sticky, where will diffusion take longer, what will be sort of AGI-Ris-Risistant or ASI-resistant for a variety of reasons, even if they're completely irrational. and you could get a better thing from a computer, but people like the human version or whatever.

The irony is that George Hatz, hitting the timeline to raise money, makes me more bullish on acceleration. Yeah. Because if not, he's obviously not historically been a huge fan of venture capital. Well, he's not raising from traditional VCs, he said. He specifically said he wants to raise from like basically like friends and family type of round. He does want 20 million.

We have this in the timeline somewhere. Where is this post from George? I'll pull it up if you want to keep. Tiny Corp. He said that he's raising $20 million to buy a building. He's going to buy a building. Here it is. So he said, this is from Tiny Corp, the makers of Tiny Grad and the Tiny Box.

If Tiny Corp was raising $20 million at $200 million valuation, who'd be interested? Business model is basically this. By an $11.5 million building with five megawatts of power, link in our Discord, wait for AMD to launch the RDNA 5 96 gigabyte cards mid-27, pre-order 3,000 cards. Hopefully we can negotiate for $2,500 each, build $520,000 tiny boxes with six of the cards in each box, run all the Chinese LLMs, make $600,000 per box, month revenue selling tokens on OpenRouter. Market depth is there. This is 1% of OpenRouter. Improvements to TinyGrad yield revenue improvements due to how power is priced in Oregon. It's only like $50,000 for the electric bill before the four megawatts, before they price for peak, not usage. We get like 3C kilowatt hour power, three cents per kilowatt hour. We can also make $100,000

per month leasing co-location space to comma. Building and car. paid off in three years max, investment made back, low risk of being undercut since we're using consumer GPUs and running the cheapest KOLO you can believe. If someone chill wants in, I'd do it. I'm not going to fake, I'm not going to hype fake tech, but demand for tokens is going to skyrocket. Look at the open claw install numbers. With crazy good optimizations, we could potentially get 3x more from the machines, and we have electricity for 3x more machines, 5.4 revenue per month, then continue to scale from there, custom chips, etc. He's starting a neocloud, or he's starting, yeah, he's going to be serving tokens.

So that was, that was, you know, incredibly bullish. You know, demand is definitely there. At the same time, he sort of took the other side of like, this is different. This is crazy. He says there's no AGI. There's no magic threshold. You guys see auto research change the random seed from 42 to 137. And, oh, M.G, it's AGI.

It's over. Yet you critique the junior engineer for the same stuff. The cost of development is falling. Overpaid engineering struggles to compete. That's the story. And he says, humanity has been recursively self-improving for centuries. So everything new is old. No, obviously I have a huge amount of respect for George.

But at the same time, saying there is no artificial general intelligence when you're also arguing that engineers are struggling to compete with new technology. that is replacing them. How is this not artificial intelligence competing with? Yeah. So I think he is agreeing with that, but it's that there is an immense amount of desire for this binary moment.

This is the singular. This is H-G. This is A-SI. The RSI is here. This thing is happening right now, and there's before and after, and everything has changed. And he just doesn't see it that way, I think.

I think he sees it much more like the internet, the mobile phone, like other technologies that have been rolled out, and electricity. Yes, there is like a before and after, but you can only really, you know, define the period by maybe a decade, and you need a few decades to understand that moment. It's very hard to go back and, you know,

there is like the iPhone moment and there is like, you know, the first launch of, I don't know, AOL. Like I don't even know, I don't even know what the iPhone moment of the internet was, just because it was sort of a slow rollout. Anyway, live GPU clusters. Where's Jordan going? Oh, you're moving the goalposts. Okay.

Jordy's moving the goalpost. Where are you moving the goalpost to? Just over here. Okay, but why? Is there a reason? Well, I mean, they have to you move. George is moving the goalpost. Oh, okay, okay, okay.

He's moving the goalposts. So I said, you know, it's clear that the AI industry continues to grow and continues to need more and more power and compute, as we've seen from George, Hots' new project. That means large data center campuses. But if they're not in random office buildings that George is picking up on the cheap, they're probably going to be built with construction equipment. So what does this mean? They don't just drop from the heavens. They require building, which requires oil, but how much oil? And is oil a serious, is a serious oil shock enough to impact

the AI buildout in a meaningful way? Spoilerly, basically no. But live GPU clusters in the United States do not use much oil directly. Only 0.6% of US electricity in 2024 came from petroleum. We're much, much more dependent on natural gas, something like 42% of U.S. electricity is natural gas. And so America ramped up natural gas production significantly over the past two decades. A lot of that was in reaction to the wars in the Middle East. Hey, we need to be less dependent on foreign oil.

We need to be energy independent, and so you have the fracking boom, the natural gas boom, and that's where a lot of our fossil fuels come from today, I believe. Data centers only consume a single digit percent of U.S. electricity. So you're looking at 0.6 percent of a few percent is like the actual impact. So the short-term impact of high oil prices should be very limited on AI. Building out new capacity is much more. That said, it's worth noting. I think it's something like Qatar produces something like,

something like 20% of the world's liquefied natural gas, right? We're not for us. Yeah, not for us. So we're in a good spot, but this is going to impact the data center buildouts or active data centers across the rest of the world if this continues. Yeah. So when you're talking about building new capacity, building new data centers, oil is a little bit more involved.

So diesel, powers, trucks, trains, boats, barges, generators, pumps, compressors, excavators, and tons more construction equipment. Petroleum is also broadly used for plastics, polyurethane, and solvents that all work their way into the data center supply chain. The biggest problem is delaying already tight schedules because of narrowly available components going out of stock. So the price of oil goes up. There's one marginal factory that can't produce one ingredient that goes into the rack and that slows things down. and you have to wait a week while you find another supplier, that stuff can add up to just a little bit of a delay.

This happened during COVID and the AI industry was already experiencing something similar with transformers. And so you don't want products getting stuck in transit or going out of stock. But the bigger problem and the one that people should be talking about, and I think you were debating with Dan Primmack at Axios about this is macroeconomics. So higher prices, higher oil prices lead to higher inflation.

If the Fed has to raise rates to control inflation, capital formation for mega projects gets a lot harder. So JLL has this estimate, the next 100 gigawatts of data center capacity could require about $870 billion of new debt financing. And so using this rough number, every extra 50 basis points of borrowing cost on 870 billion is $4.35 billion in annual interest expense. And so that's a, I don't know if that's a huge squeeze on cash flow, but it's certainly a squeeze on cash flow.

if you're a hyperscaler and you're going gig along AI CapEx now, and you're like, yeah, we're going to do 200 billion this year, we're going to do 500 billion next year or whatever across a couple companies, you start looking at interest expense bills that are already huge, but then they're going to go up by $5 billion or up by a full extra billion just with a 50 basis point increase, half a percent. So half a percent change in the Fed funds rate can echo through, Even a quarter percent could probably be amplified into a multi-billion dollar cost line on your balance sheet or on your income statement as you're servicing that debt.

Overall, higher oil prices do make the AI rollout harder, but not really for existing AI capacity. I wouldn't expect token pricing to change based on the price of oil. But there is this bigger question of macroeconomic resilience during a time where our largest tech companies are digging through the couch cushions to find every penny. to win the AI future. Middle Eastern investors pulling back, I agree with you. I do think that this is a significant risk because if they pull back on big investments makes these mega rounds more difficult,

retail investors could fly to oil stocks or defense stocks. They think that this is like going to be a thing for months and months and months. It could all flip if there's a quick resolution though, which is what I'm hoping for. But if things continue, the AI industry walks an even tighter, tightrope. Yeah, I mean the big question right now is the hundreds of billions of dollars

of sovereign AI projects in the Middle East, right? I think a lot of those, people are going to be like, do we want to send, you know, billions of dollars of GPUs? Over there, and then also the money coming here is another thing where you might want to spend it elsewhere. Quickly, let me tell you about Turinwell Puffer, serverless vector and full-tech search, built from first principles in object storage, fast,

10x cheaper, and extremely scalable. And let me also tell you about public investing for those who take it seriously. Stocks, options, bonds, crypto, treasuries, and more with great customer service, whatever investing strategy is. And we have to say, happy birthday to, Trey. Happy birthday, Trey.

We love you. I'm very excited to have you. Thank you. Of our strongest soldiers. We appreciate you. It's been an honor podcast with you. Have you been noticing that it's been hotter in Los Angeles? I have.

Downtown Los Angeles is forecasted to approach a hundred degrees Fahrenheit on Thursday and Friday in March. Which is why we're going to do the weather. We have a new segment for you today. We have a new segment for you today. We're doing the weather. the weather on TBPN. We have our very own Ben. We have Ben. Hello guys. How are you? We're doing great. Tell us about the weather. What's happening?

Well, I want to start off by saying, as you can see, the weather today, for the low of today, is going to be 75 degrees Fahrenheit up there, high of 100 degrees Fahrenheit down there. But there's something I actually wanted to point out that I saw and I thought that was quite interesting. As you can see up here, there's a localized low pressure area up there and a localized high pressure area right down there, if you can see that right there. Does that mean rain? The issue, that's not normally an issue and not a cause for concern, and it's not very common for this time of the year. However, today, later in the afternoon, these two areas are going to collide and they're going to hit each other. Really? And what that's going to cause is a barometric pressure inversion.

It might sound a little bit scary, but I guarantee there's no cause for concern. All that means is that hot air rushing in from the west is going to collide with that cold air rushing in from the east, and it's going to cause a bunch of turbulence in the sky, moving all the airwaves around and oscillations in the sky. However, I want to add one more point. A byproduct of this effect is that all that humidity that dropped after that hot air moved to the bottom is going to raise up because the water cycle, you know, evaporation and stuff. It's going to raise up into the sky into those clouds. It's going to cause big clouds in the sky.

And eventually all that water is going to fall down onto the ground. We're going to have big rain later in the afternoon. It's actually going to rain in L.A.? Yeah. It might sound crazy, but I just want it for all you guys at home, I definitely try to step outside with a jacket today, maybe a hoodie just in case the rain comes. Don't try to be a big shot. I'm fact-checking you right now,

and the weather app says it's going to be sunny all week. Is this just complete fake news? Those are the numbers my team gave me. This is complete fake news. This is the fakesest news I've ever heard. Literally, the transatlantic current. John, John, I don't trust your app. I trust the weather.

Apple says it's not going to rain the entire month. There's zero chance of rain. Did you look at the transatlantic current? No. The transatlantic current. Precipitation. Zero inches today. Zero inches tomorrow.

Zero inches on Wednesday. John, you're really going to trust an application. You're really going to trust an application. Yeah. It was probably vibe coded yesterday over Ben who's doing the weather. How did he get here? What happened here? Yeah.

This is a no. If you look right in there, you can see a localized high pressure area. That's a localized high turbulence. I mean, this is an over-eager weatherman who's just looking for drama in the most boring weather market in America? It's Los Angeles. I think, I think this is the most important story in the world. This is ridiculous. You guys can look at the jet streams. They're coming in from the west. I don't want to hear any more mumbo-jumbo about jet streams. Get out of here. You're done. Great work, Ben. Thank you, Ben. The fake news weather will return. Maybe never. Who knows?

Let me tell you about Restream. One live stream 30 plus destinations. If you want a multi-stream, go to Restream.com, and let me also tell you about Gemini 3.1 Pro. With a more capable baseline, it's great for super complex tasks like visualizing difficult concepts, synthesizing data into a single view, or bringing creative projects to live. This is the future of the weather. You have a weatherman who gets into a live altercation with one of the other host, who's just constantly fact-checking. It's on site next time, Ben. It's on site. If you pull up with some fake news on TVPN's weather report, it's on site. Ben, you crushed it. Good job. So it is, it's not Adam Smith's birthday.

It's the 25th anniversary of the publication. 250th, sorry, 250th anniversary of the publication of the wealth of nations. Tyler, you have a copy? I have two of our copies here. You have two of our copies. So, yeah, usually I have one with me, but I didn't bring mine to work because we have a, we have a few years. Yeah, yeah, yeah, yeah, obviously. For those who don't know, give us the pitch for wealth of nations.

Why should people read wealth of nations in 2026? Yeah, I mean, this is kind of the OG, you know, economics book, right? It was, like, put it into, like, AI terms. This is, like, situational awareness type of, or not, that's, like, this is like, attention is all you need. It was, like, the chat, GPT moment for capitalism. Yes, yes. But I would like to, you know, I have a few quotes I would like to read.

Okay. Let me open them. Let's see. It's a big book. It's like 800 pages. And it's really, really dry. I mean, it reads like, you know. was written 250 years ago.

Okay, so this is, I think this is kind of one of my favorite quotes. Yeah. Kind of the, you know, seminal. Every individual necessarily labors to render the annual revenue of the society as great as he can. He generally, indeed, neither intends to promote public interest, nor knows how much he is promoting it. Yes. He intends only his own gain. Yes.

And he is in this, as in many other cases, led by an invisible hand to promote an end, which was no part of his intention. Yeah. Rational economic actors, homo-economicus, thinking of the value for society is created by rational self-interest. Sort of the spin on like greed is good.

He was like an OG of that thesis. Greed maxer. Yeah, productive labor, efficient markets, free markets, don't hold gold, don't control trade. An important read in 2026. Somewhat heretical. There are many people that

disregard that book. Maybe they should not. Maybe they should give it a crack. Evergreen. I think it's evergreen. I think it's ever green. I mean, there was that debate over Tyler Cowan over the weekend, and somebody was taking shots at stubborn attachments for being, like, not advancing the discipline of economics. And my pushback to that was that I actually think it is an incredibly, it's just a fun read. It's like really well written. It's very entertaining. and it's very thoughtful. But it's particularly interesting to hear someone sort of just rehash

the tenets of liberal democratic capitalism in 2020, I think you read it in 23, 2022, something like that. But like, there are a lot of voices out there who are anti-capitalist, anti-democratic, anti-liberal, and Tyler Cowan is just retreating to like a time-tested, time-on- economic philosophy that is deliberately not revolutionary.

And I thought that it was a great book for that reason, and that's why I recommend stubborn attachments to so many people and so widely. But if you want to crack open the real hard stuff, pick up a copy of the wealth of nations. Why is one of those copies so much smaller? Is the font size just smaller? Yeah, it's super small.

It's also two columns. Oh, two columns. Wow. Yeah, if you can get through that, you're truly, you're truly elite in the global economy. What does the timeline have to say about the masterwork that is turning 250? It's now needed more than ever, says the Washington Post. On March 9th, 1776, four months before the American colonies broke with Britain over the issue of taxation, a little-known Scottish thinker published a long, dense book with an unpromising title. Shops fire. An inquiry into the nature and causes of the wealth of nations. 250 years later,

the Adam Smith is, by any objective measure, easily the most widely cited and widely quoted economist who ever lived. Astonishingly, his work still frames the central questions we face, not just about free markets, trade, and capitalism, but about the nature of human society itself and even what it is to be human at all. Smith was born in 1723 in Kirkcaldy, Scotland, educated at the universities of Glasgow and Oxford. He initially made his name, not as an economist, but as a moral philosopher. His first published book, The Theory of Moral Sentiments, offered a radical theory of how we form moral judgments, radical because it derived from the creation of moral values, not from scripture or divine

grace, but from human sympathy in mutual regard. The wealth of nations, as his second work. Second major work came to be known. It was an extension of that product. The book is not as sometimes believed, a hymn to greed, a pain to market fundamentalism or red in the tooth claw capitalism. It was an attempt to understand how a commercial society could generate prosperity without collapsing into corruption. This 250th anniversary is not a moment for hagiography. It is an opportunity to recover a way of thinking that is directly relevant, indeed urgent to the economic, social, and political challenges we face today. Begin with trade.

In his own time, Smith's great target was mercantilism, or what he called the mercantile system, the idea that wealth consists of hoarded bullion and that trade is a zero-sum contest. Governments granted monopolies, imposed tariffs, and manipulated commerce in the purported pursuit of national strength. Producers were widely protected. Consumers often ignored altogether against this. One of these days, we should have Tyler just do a speed reading of the wealth of nations.

live on the show as long as speed run uh yeah wealth of nation speed run any percent glitchless you need to do glitchless though for sure a glitchless run is the only way we can do it no glitching uh no no ai summaries i think that would take you probably five days probably five days it's so dense have you seen how fast you can rip a rubic cube this edition is 488 pages around oh yeah I guess you don't have to read all the footnotes and stuff. You can kind of skip through it, but it is a thick book. But the way it's written is just really not fun to read. It's like it reads like it's 250 years old.

It's not like a page turner. Well, we need to talk about the wealth. But first, vibe.com. We're DDC brands, B2B startups, and AI companies advertise on streaming TV. Pick channels, target audiences, and measure sales just like on meta. And let's also tell you about fin.a. the number one AI agent for customer service. If you want AI to handle your customer support, go to fin.a.

We need to talk about the wealth of Sundar Pitchi. Pitch AI. A new pay deal worth up to $692 million. Is this like 10 times what Tim Cook's making? No, it must be over time, right? I mean, Cook's making around 70 a year. 70 a year. Combined.

So if he works for 10 years, he makes what Sundar makes in three? there's going to be a conversation. I mean, but this is what we've been we've been advocating for this. So this is good. Yeah, no, no. We're, we're, we're, we're, we're, we're, we're, we're, we're, we're, we're, we're, we're, we're, we're, we're, we're,

you know, you know, you know, you know, you know, that Tim Cook dropped this in the Apple board members group chat as soon as it hit. He was just like, dude, this is the, this is a cool article. Yeah, just drops it in just drops it in lots. Yeah, yeah, exactly. The bulk of his package comes in performance units. with a target value of $126 million split evenly into two branches. The PSUs are valued by the parent company, blah, blah, blah.

It could pay out as much as twice the target, a quarter billion if it outperforms significantly or nothing if it lags behind. Got to beat the S&P 100. Pichai will receive Waymo stock with a target, stock in Waymo with a target value of 130 million and 45 million in wing aviation. That's their their germ delay. own delivery platform. Platform. Again, both can pay out up to 200% of the target. They've got a lot of incentive pay.

Like if he delivers and the company does well, he should be richly rewarded. Barron's was writing about the death of the Mag 7. Mag 7's over. Up and down Wall Street, the Mag 7 era ends. Barron's is calling it. They said, stick a fork in it. Turn out the lights. Asta la vista.

Say it anyway you'd like. The simple truth is that the Mag 7 trade is over. Finito. I love Barron's rating. Dead. The collective stock market outperformance of those seven tech icons, Alphabet, Amazon.com. Apple, meta platforms,

NVIDIA, Microsoft, and Tesla is now a thing of the past. The group may still do okay, and some of the individual stocks may even kill it, but the slam dunk set it and forget it, run circles around the market era of the mag seven is gone with the wind. Recall that B of A securities analyst Michael Harnett coined the term mag seven in 23, pretty recently, referring to the 1970s John Sturge's western gunslinger flick, an adaptation of Akira Kurosawa's 1954 film seven samurai. Have you seen seven samurai? No. Have you seen the magnificent seven? The stocks?

No, the movie, the movie, the western. No. He's done those seven samurai. In what world? Production team. Has anyone seen either of those movies? I think we watched them in school. Okay, okay.

You're saying you didn't watch them, you studied them. Well, yeah, yeah, yeah. I haven't seen them in a while. Which ones have you seen, Tyler? Seven Samurai. Seven Samurai. It's a great one. It's a very good movie.

Anyway, Barrens is bearish on the Mag 7, but Seardley Chide's Dublin. It does sound like cope. I think the Mag 7 will do. It's like the large, it's, for the largest and most profitable companies in history that stand to benefit some ways the most. The actual argument was the growth to value narrative, the loss of cash flow as they increasingly invest in CapEx. The financials will look very different. There'll be margin compression,

that type of thing. It's not an unreasonable take, but it is just funny the way it's written. Anyway, Sundar Pachai took over his CEO in August of 2015. He's going on 11 years in the seat. Google's market cap has increased almost seven-fold from half a trillion to $3.6 trillion, briefly topping $4 trillion in January. This surge has made the Indian-born 53-year-old former McKinsey consultant, a billionaire. He joined in 2004 and made his name developing the Chrome browser and leading the Android division. He had been criticized for being too slow to adopt AI at the search giant, allowing OpenAI to release the first hit product, ChatGPT, in late 2022, but has since been,

bounced back, releasing cutting-edge AI models and integrating the technology into its dominant search engine. Yes, he's done very, very well on that front. Pachai has also navigated a duo of antitrust cases brought against Google's search and app store businesses avoiding the worst-case scenario of forced breakup. A third lawsuit is pending against the advertising network. Tyler, do you think that DeepMind counts as RSI capable at this point? Recursive self-improvement?

So there was a take on the timeline where basically someone was making the argument that XAI is cooked because while Anthropic is using Claude to build Claude and OpenAI is using Codex to build Codex. So both of those leading private AI labs are recursively self-improving. XAI was using Claude code to build GROC. And so they don't have the same flywheel of the tools that build the tools, build the tools. Yeah, because I think they lost access to cloud code. Yes. But then they still have access to codex. Okay.

So you can, it's kind of like, do you count that as RSI? Because it's not like really internal. Yeah. But is Gemini? That's the question. Yeah, yeah. I'm sure people at Gemini are using Gemini. I mean, there is a Gemini CLI.

Yeah. People don't talk about it as much. Yeah, yeah, yeah. I mean, the model is very good, right? Yeah. Yeah. I assume it is. It would be interesting.

I feel like we don't get as many little data points from the Gemini and DeepMine team because it's a public company and they need. need to, you know, maybe break out financials at some point. Yeah, I think like vague posting is, is probably, you know, more heavily looked at. Exactly, exactly. Um, because you would, yeah, because it's totally possible that they're doing the same thing. I mean, Demis has been extremely, uh, just very clear about the path this goes and he's very AGI-I-pilled. He understands the exponentials and whatnot. So, uh, he's certainly messaged it at a high level without actually saying, you know, definitively, yes. Like, we are at a point where we're,

not writing any code by hand, that type of phrasing that you see in the RSI crowd. Anyway, Pichai last got a stock award in December of 2022 worth 218 million, which was structured in the same way. His earnings are topped up by his personal security costs, which rose to $8.3 million in 2024. Earlier in the week, he sold 32,500 Class C shares, an average price of $303 worth roughly $10 million. The Bloomberg Billionaires Index estimates that he has sold about $650 million in stock since becoming CEO.

They still own, along with his wife, he owns 1.67 million shares of Google worth half a billion at the latest stock price. And Google's founders, Sergei Bryn and Larry Page still control the company through their ownership of super voting class B stock, which gives them 56% of decision-making power. interesting. I think you've got to give Miami... Paper hands? He sold 650 million of stock? No, no, no. He's sold that, but he's... Oh, and he only owns half a billion. So, yeah, he's sold more than that. I'm just saying, hopefully, hopefully he put it all into video. Paper hands, but he's getting topped up. And he's diamond handings the new...

I know, I'm just saying, looking at the stock chart since 2015 when he became CEO. would have done pretty well just to not do anything and trust his own process. Hey, hey, maybe he has an amazing car collection or something, you know? There's a lot of things. You can't rotate into metal, maybe. You can't drive a Google Class A share. You can't sleep on a bed of Class C shares, Jordi. You got to get liquid.

No, you got to give Miami Tech it's due because everyone's saying, oh, Miami Tech, they don't have any startups. There's no funding there. Miami Tech isn't panning out. Well, Google is now counts as a Miami tech company because the Google founders bought houses there. You also got to count meta as a Miami tech company because Mark Zuckerberg has a house there. So you put meta and Google. That's like $6 trillion in market cap for tech company. All in Miami. I count it. I count it. Anyway, moving. Owner of Forloko is exploring a sale of a story to alcohol brand. Sources say.

Wow. Let's dig into this. But first, let me tell you. you about Plaid. Plaid powers the apps you use to spend, save, borrow, and invest, securely connecting bank accounts to move money, fight fraud, and improve lending now with AI. And let me also tell you about cognition. They're the makers of Devon, the AI software engineer, crush your backlog with your personal AI engineering team. So... The parent company of Forloco, the canned alcoholic beverage that became a college campus sensation in the late 2000s before being reformulated under regulatory pressure is exploring. I cannot believe they nerfed Four Loco.

It was too powerful. It was too powerful. You can just imagine the trajectory of the United States if it hadn't been nerved. Straight downward. Really? Yes. It was so bad. It was so.

I thought people were having like heart attacks. Yes, it was terrible. Oh, really? For Loco was a 24 ounce can. So twice the size of a normal alcoholic beverage. Trey, the birthday boy says it was amazing. It was amazing. Okay.

Now we have buzz balls. That's true. So for loco, the original formulation was twice as big as a normal can of Bud Light or Miller Light or Coors Light, your normal beer, something that you would grab. So it was two drinks that way. And then instead of it being 3.2% alcohol, or 4.5% alcohol, it was like 10% alcohol. It was the strength of like wine almost. And so a single ForLoco was like four or five beers in one can. hand. And then they also added like 200 milligrams of caffeine. So you would, you would become

incredibly intoxicated and inebriated, but then also incredibly stimulated from all the caffeine. And that spelled doom for many people. People would be very high functioning, but completely inebriated and discombobulated. And so they would, you know, get into all sorts of trouble, whether or not it was behind a motor vehicle. A discombobulator. It was the, it was the discombobulator. It was also wildly illegal from an FDA perspective. You can't mix alcohol and caffeine in a single product.

That's just a rule that they chose not to follow, and they figured out how to make it. And then the FDA eventually gave them a warning letter, and they pulled the caffeine out, and that, of course, killed the viral sensation that was for Loco. But apparently, the company continued to sell products that were compliant with the current FDA regulation. NERF. That's nerfed. It's the NERF for Loco continued to... Somehow they're looking to... And potentially will capture somewhere around 400 million on the sale. The brand value. So people... It's hilarious that they're working with J.P. Morgan on this. That's wild.

So Fusion Products is the owner of the brand. They're working with J.P. Morgan on the sale process. The potential sale underscores how ready-to-drink beverages have emerged as a growth category. an otherwise sluggish alcohol market. U.S. beer and wine, wine, and spirit sales declined in 2025. We've talked to a number of guests about the declining alcohol sales as folks move over to more functional beverages. Here's what the CMO had to say. For over a decade, our sales have been the leading flavored malted beverage have been leading the flavored malted beverage market by embracing bold innovation, unconventional marketing, and a risk-taking attitude that delivers results a

after year. I guess they kept building. They also own... The parent company also owns Mubesas, pirate water. Pirate water. Basico tequila. What is pirate water?

And just earthquake. Earthquake. Earthquake seems like something you definitely want... From the makers of Four Locco, earthquake. Pirate water is a wild name for a brand too.

I mean, it's a distinctive can. It looks remarkable. It jumps off the shelves. When these, when the Four Loco Nerf went into effect and they pulled them from the shelves because before they reformulated without the caffeine, they had to pull them from the shelves entirely. They were a hot commodity and folks would go around stockpiling them before they were banned and they were one of the most guarded resources in all of college campuses. Well, earthquake is just a can that's just 10% alcohol. Wow. Do not recommend.

Very risky. Never quake. Anyway, in 2010, the FDA sent warning letters to caffeinated alcohol beverage makers, including for a loco, saying that the caffeine in their products constituted an unsafe food additive. So the FDA has specific categories for regulation around what is an alcoholic beverage, what is a caffeinated beverage, and they never the two shall mix. So they don't let you put these two things together. Of course, vodka red bowl was the precursor to this where folks would order a red bowl and then they would put vodka in it, I think, but at a bar, which is made at the time of sale, and so that was legal, I believe.

And this might be fake news, but I remember hearing a story that at one point, Red Bull was sued because a gentleman had purchased a series of vodka Red Bulls, become incredibly inebriated, gotten behind the wheel of a car, gotten into some sort of accident and wound up in front of a judge. And the argument was that he wanted Red Bulls, to be held liable for his actions while inebriated behind the motor vehicle.

And his argument was that it was only because of the product, Red Bull, had been mixed with the vodka, that he made the decision that he was alert enough, stimulated enough from the caffeine to get behind the car wheel in the first place. And this might be apocryphal, but the story goes that Red Bull made the argument that the caffeine actually made him more capable to drive, not less capable to drive, and actually increased the safety of the driver and the vehicle. And so they were held not responsible. But I don't know. That's just like an old wives tale on CPG.

I don't know if it's true. Anyway, let's move on to Phantom Cash, fund your wallet without exchanges or middlemen, and spend with the Phantom card. And let me also tell you about Label Box, RL environments, voice, robotics, evals, and expert human data, label box is the data factory behind the world's leading AI teams. So we're all preparing for the singularity. What is this doing dishes? Just like, oh, if you're doing dishes, you're not worried about Dylan Patel's, you know, viral post saying that being in SF is like being in Wuhan right before the pandemic, something is happening. It's going to hit everywhere. But so few people know it.

And that is the question of, are we in a software-only singularity where the dishwasher is life as normal? and every time I go to the COVID analogy, I start a clock because I remember tracking COVID in January, and I remember the viral post, biology posted in February, going viral. And then in March, the NBA shut down and in April, everyone was locked down. And so every time I see one of these, it's like COVID, I'm like, okay, let's see what it's like in eight weeks. Like eight weeks is not that long. In eight weeks, if I go outside and. things look radically different and you talk to random people and it's wildly different.

Then we're like, okay, yes, that analogy was correct. But if it's more like in eight weeks, we're like, oh, yeah, like, you know, technology is getting more exciting and, like, companies are changing and, like, the market's moving. But, like, you know, the dishwasher is unaffected. The pizza place on the corner is unaffected. That's always how I, I just think people underrate, like, how crazy a global plague was. Like, it's literally like a once and a hundred. 100, like the most insane thing ever.

It's the craziest change to literally everything from the sports that you watch. You couldn't go to the movie theater. You couldn't go to dinner. Like, everything was changed for like a full year. It was like complete upheaval in our society. And it might happen. But eight weeks, eight weeks is a little fast for me. And if you live in L.A., probably still be able to go to the beach.

I think so. So that's... But maybe not. Maybe everything's... I did have a funny moment yesterday. I was telling you, but I'll share it again. I was talking with my car detailer who's become a dear friend, and he was telling me that the people in my area cancel their appointments at a much higher rate than the people in the rest of L.A.

And he was talking about running analytics, I messaged, and I was like, how do you do that? And he's like, he's fully running OpenClaw. So he's now using OpenClaw to run his... like, car detailing business. He's like, yeah, just always running in the background. It's, to him, he was like at some combination of having, like, an EA or a CEO, like helping me manage a calendar, helping with comms, helping me make smarter decisions on routing and scheduling and all these things.

So when you book a detailing, meaning that this gentleman is coming to your house and cleaning your car, you send an I message to him, correct? And that is what is so interesting about OpenClaught, is that there were probably a thousand companies that were like, we're vertical SaaS for auto detailing. You got to do everything over email. And they just got eaten alive because people use iMessage. And iMessage has been this closed ecosystem that private companies could never access before. But it makes so much sense for him to say, hey, run through all my I messages. and if I'm talking to my friends about the F1 race,

just discard that. But if I'm talking to a client about potentially booking a detailing, put that in a spreadsheet, organize that, and then run some analytics and tell me what the patterns are. And that's just something that should have been SaaS for a decade, but Apple has a walled garden that has now been smashed down by OpenClaw. And I think it's really cool.

And I think that that's what we're seeing, the first use cases for open call. It's like things that could not happen for business reasons, for logical reasons, sometimes for legal reasons, if you're talking about downloading illegal films like the Napster analogy. But a lot of times it's just Apple didn't want to do a deal to integrate iMessage into Salesforce. The way to break down the walled garden was just to have a digital guy. Basically. Yeah, who just, who just combs over all this. So is that a new type of software that should have existed that now is unlocked? Is it disruptive?

It's not like he stopped using his previous CRM probably didn't have anything. So now he has something? Yeah, he's pretty tech forward. He's tried every single SaaS tool that there is. But like at the end of the day, having like a digital guy that has access to all of your tools is just a much better experience.

And so I want to know what his bill is. He said he spent a decent amount. I want to know what it is this month, and then I want to know what it is in like a year, because sometimes the maintenance of these systems can be really big, and then there's a question about, are we going to see per token deflation, or are we going to see prices? Are prices being artificially held low because of competitive dynamics? So in a year, will he be spending more or less?

Will he still be getting the value? These are all interesting questions, but I'm just excited that a digital product now exists that could not or didn't exist before, and that feels incredibly positive and cool. So very exciting. Anyway, we have our first guest of the show in the Restream waiting room. Let me tell you about console.com. Console built AI agents, though, I mean 70% of IT, HR and finance support, giving employees instant

resolution for access requests and password resets. And without further ado, Dr. Alex Wisner Gross, welcome to the show. How are you doing? It's happening. It's doing really well. Thank you, John. Thank you, Jordy. Thank you so much for taking the time to come chat with us.

congratulations on all the virality. For those who haven't followed along so far, can you introduce yourself and the project broadly? And then I have a ton of questions, but I'll let you kick it off with an intro. Absolutely. So quick background on myself, originally from New York,

undergrad MIT, studied physics, electrical engineering, computer science math, PhD physics, Harvard. Was there anything you didn't study? You studied everything. Didn't study the humanities. Okay. Consider going to Cambridge to study chemistry and biology

for a while, disbarred it. I was told physics as a game for the young. So, stayed in the U.S., did my PhD in physics, focusing on AI and nanotech. I've started, invested in, advised, managed dozens of companies at this point. I've had a number of exits. And at this point, most of my time is focused on smoothing out the singularity

as you guys were just discussing. So most recently, as you were gesturing a company that I helped found, Eon Systems, Public Benefit Corporation, just announced what we've been characterizing as the first multi-behavior upload, if you will, more on that in the second, of a fruit fly, which we think is a major step forward for the field.

And quite frankly, in an era when I do another podcast, the Moonshots podcast with my friend, Peter, I talk all the time about tiling the Earth with compute. Right now, all of this AI infra buildout that we're doing to the tunes of trillions of dollars of CAPEX, all of this is going to artificial minds. and Eon is playing, I think, a very important role in doing some early pioneering experiments and developments to try to level the playing field of the future, not just artificial minds playing on all of this trillions of dollars of infra and data centers, but enabling emulations and, if you will, uploads of human minds, non-human animal minds, at the moment, starting with fruit flies and aspirationally working towards mice and humans. getting the brains of a variety of human and non-human animals to operate in the cloud. So walk us through the fruit fly experiment, why it was so exciting for you? It's really interesting. If you think back, remember back with the launch of Chad GPT,

that was just a side experiment. That wasn't the main focus of OpenAI at the time. Some have characterized that as an unhobbling, if you will. GPT3, the leadership of OpenAI thought at the time was, sort of the main entree, if you will. Similarly, within Eon, I don't mind saying this was more of a side project that we thought it would be somewhat interesting to the world, but I think the level of response and virality probably even took most of Eon by surprise. So this is a video demonstration of an internal project to take a bunch of building blocks

that were already sitting around for the past one to two years and put them together for the first time. So one of these was a paper that senior scientist at Eon, Philip Chu, published in Nature in 2024, demonstrating that just from an emulation perspective that you could take the flywire, which was a well-funded project to capture the connectome of the fruit-fly brain, that you could emulate certain circuits within the fruit-fly brain. Fruit-fly brain. We've had that for a year and a half or so.

Another project, Neuro-Mecfly version 2, is an excellent project to be able to formulate simulations in realistic mechanical environments of fruit flies. Another project from Erzdel et al to formulate coordinated motor actuation of fruit flies. And then AI. So in summary, Eon was really the first to put all these building blocks together that we're sort of lying around. This is actually a tiny subset of what Eon is working on. Eon is working on scanning the connectomes of human and non-human animals at scale and achieving whole-brain emulation, which is the Holy Grail of the field.

There have been many projects in the past that have attempted human whole-brain emulation, Eon is leveraging advances in AI, expansion microscopy, a variety of other techniques. But it was really just putting the pieces together. So we're quite honestly astonished by the level of interest in just this one demo. I love it. I have a million questions. Let's start with I.O.

Like the inputs and outputs, like, what are you simulating? When I pull up the Sims, that is somewhat of a simulation of a human, you know, a basic for loop over a bunch of random probabilities to say, are you hungry? If hungry, eat, go over to the stove, cook food. That's obviously nowhere near the level of detail at which a human operates. Is there an abstraction layer here where the output is quantized to things that you would think a fly would do, like move a wing? Or are you operating at the level of like choose to fly? Or how, at what level of abstraction is the fly's output happening?

Yeah, this is what I was gesturing at a moment ago. So it was a paper by Ersdell at all that bundles most. motor actuations in terms of high-level representations. Got it. So a lot of people, sort of the people who got perhaps most excited in the neuroscience community about this video, some of them I think didn't bother to read Erzdell's excellent paper. So a lot of those, this is why I'm saying a lot of the pieces were sort of lying on the ground, waiting for us to pick up and put together.

So these abstract motor movements were already available. Just no one had ever prior to this bothered to wire them on. That's cool. Then what is the do you need a system prompt? Like do you need like an like an initialization function or once you once you synthesize the connectome synthesize the connections in the brain? Things just start firing. Do you have to trigger something to to to start the process of consciousness or whatever you call whatever the fly is doing? No system prompt needed. On the other hand in the early days of chat GPT before instruction tuning, there was no system prompt either. In the days of early language models,

no one had thought of prompt engineering. So it's entirely possible that maybe in the future, some sort of prompt engineering equivalent for whole brain emulation will be essential, but at the moment, there's no direct analog of that.

Help me compare some of the scales that we're talking about on complexity of a fly, complexity of a mouse, complexity of a human. I imagine that we're on some sort of exponential here, but how many orders of magnitude in complexity, size, number of neurons, something like that, whatever the metric is, how far away are we from humanity? We're many orders of magnitude away. So the canonical estimate for the human brain is, depending on how you count brain cells, and there are multiple types of

brain cells. Call it order of magnitude 100 to 200 billion cells. in the brain were many, many orders of magnitude, many orders of magnitude away from that. So this isn't going to happen immediately. We're not going to get human whole brain emulation immediately. On the other hand, I think it's important. I think back to the early days of driverless cars and autonomous vehicles. And what the world is missing right now is, I think, sort of a scale or a framework for starting to think about what that future looks like.

We're at the earliest stages of that, call it level, or level two uploading. A lot of people even get triggered by the term uploading. They prefer something else. And I think we need to start having these conversations in a critical, thoughtful way and start to define what is a framework for multiple levels of fidelity

for uploading or emulation or maybe we come up with a new term. But right now we're starting that discussion and we're trying to do it in a thoughtful, responsible way. So you like the term uploading because you did not choose a, random structure or an average structure of this particular flies brain you chose the exact structure of a particular flies brain this is a very expensive structure of a very particular flies fly's brain

this was that the structure was again captured through the flywire project at great expense so that's like a clone digital twin type of thing right that is that aspirationally i mean again we're at the earliest stages i don't want to oversell yeah this is just a protocol And quite frankly, this is a sliver of everything that Eon is doing. But the fact that Eon was able to put a few of these pieces together, build a prototype with relatively small effort, suggests that, you know, we talk in the AI community all the time about overhangs. Many people thought that the arrival of large language models

was, in some sense, a technological overhang. We could have, one could imagine a thought experiment? Could we have had large language models 20 years ago if we simply knew what we should have been building? I think probably... I thought we were very GPU-gated on that. Well, the GPUs were being spent on video games for the first few decades. So we could have, in principle, like, Markov models, we knew how to build Markov-Bablers.

I was building Markov-Bablers in the 90s and early 2000s. Yeah, it just feels like even if you try and port back Frontier models now to a, you know, 1080TI, which was a graphics card for gaming, you're going to have a bad time. Well, I mean, with the rise of SLMs and some of the amazing deflationary properties of algorithmic progress in SLMs, I think we will find ourselves in the world next two years where some of the earliest PCs could probably have hosted non-trivial conversations with LLMs. There's just that much algorithmic progress. Extremely bullish for Apple. Anyway, not to put it in hardware terms. What do you want to do next?

Like, where do you go from here? I know it's a... Well, more flies or straight to mouse? It's such a profoundly interesting question. So, Eon is beginning fundraising. Eon wants to tackle both mice and men. And I think the time scale will be determined by both technical capability increases and also fundraising abilities.

So I think it's an interesting time. I like to say, you know, on my other podcast, that the singularity looks like all sci-fi tropes happening everywhere all at once. And one of these sci-fi tropes is most definitely mind-uploading. And I'm doing my best vis-a-vis the Eon systems to push that part a little bit to the left so that the AIs don't have the Dyson swarm all to themselves. Interesting. Architecturally, what does the neural network look like that you've built for the fly?

Is it similar to a large language model like transformer-based architecture? Or are you a beneficiary of attention? Totally unlike the transformer. Totally unlike the transformer. So transformer architecture, if you subtract off the encoder and decoder layers, it comes in many variants, but the most vanilla variant looks like alternating linear and attention layers. So totally unlike the transformer architecture, at least if you look at it, it looks like a little bit more like a graph neural network,

but really that's such an AI way of framing it. It looks like a leaky integrate and fire, LIF model that we've had for decades from the neuroscience world. So it's just a graph of nodes that have leaky, integrate, and fire dynamics, and they're firing at each other. So what does that tell you about the nature of synthetic human intelligence? Like, do you think that there is a path to, human level AGI that does not involve the transformer

and instead falls along your path and your architecture? Oh, sure. Well, first of all, I would argue we've had AGI for at least six years now, five to six years at the very latest. Since summer of 2020, when OpenAI published their language models or a few shot learners paper, yeah, GPT3. I would argue that was an AGI. So I think there are many ways, my bet is there are many, ways to achieve generality of intelligence. I published a paper a number of years ago arguing that not only is intelligent behavior a general process, I went further and argued that it is a general

physical process that you can even formulate intelligent behavior in pure thermodynamic terms. So my bet is intelligence is this very, very general effect and lots of ways to implement it. Interesting. Do you, so do you, I mean, you've mentioned AGI six years ago, singularity, here and you're smoothing it. Do you have other sort of binary benchmarks that you're looking towards like ASI, recursive self-improvement? Are any of these terms useful to you? Oh, we're already there. I mean, we're already in the era of recursive self-improvement. All of the frontier labs are pretty public about it. At this point, I'm looking past the singularity. So I spend most of my time on bets for what the post-singularity world looks like.

So, Eon is one of those bets, another bet. I have a company, physical superintelligence, that's trying to solve all of physics with AI and doing an amazing job. I just wrote a book with Peter called Solve Everything, arguing that entire disciplines are going to get steamrolled by superintelligence. And that what matters now is what disciplines we aim the superintelligence at. Interesting. What decisions do you make in your personal life in the way that you live that are based around your beliefs, around technological progress? There are a few that come to mind.

So one is I'm hoping not to die. It would be a shame if I died sometime soon, get to miss the most exciting developments. Same. That's what gets the applause. That's great. Yeah. That's a white pill.

White pill. Very good. Brian Johnson thanks me in advance. But I think also, I mean, there are so many other angles Another is not trying to bet against the collective intelligence of the market. I've had a number of friends who thought they could outsmart all of the AI Algo traders and day trade. And when they inevitably fail, they attribute it to bad luck or something other than just fundamentally betting against progress and betting against AI intelligence. So trying not to make the mistake of betting against collective intelligence of civilization.

and AI capabilities. I also, I'm trying to make bets assuming that the AI capabilities keep increasing to the point where they're ontologically shocking and trying not to duplicate effort of frontier labs. And I think that's perhaps something I don't see enough of in the venture community. So many new startups being formed now just aren't being ambitious enough. I have startups that newly formed or recently formed that are literally trying to grow new islands and new coastlines with AI. That would have been unthinkable years ago. I have...

That's with physical AI. Or they're just dreaming them up? It's with AI reaching into the physical world to steer ocean currents and grow new islands and new coastlines. That would not have been possible a few years ago. That's now starting to become possible. I have a startup that's working on solving interspecies communication, starting with dogs. Love that. That would not have been possible a few years ago. So I spend a lot of my time thinking about what the post-singularity state of the world looks like and how to smooth that out and bring it here sooner.

Yeah, what does the post-singularity world look like where you are able to upload yourself? I assume that feels like there's a copy of Alex in the cloud and then you still exist and not in the cloud. How do you think that that interfaces? In my mind, if we wind up having the copy question, come about, that's almost a failure mode. Speaking for myself, I don't want a copy of myself. I want myself. And Hans Moravec and others have written about this.

What I would like to see in the space and what Eon is working toward, at least aspirationally, is this idea of a continuous transfer of consciousness so that it really is you. It's not a copy of you, it's not a low fidelity. Facsimile of you. It should be a better expanded version of you that's still you.

whether that looks like moving from what Eon's doing right now, with fruit flies to maybe replacing a single cell in a brain at a time, invasively or non-invasively, with a substrate-independent or substrate-migrated implementation, some variant of that, I suspect, is where all of this will go, so that you never have to worry, is it really myself or not? It will, by construction, be yourself. And I'm not sure of the precise timetable,

but I think five, ten years from now, I think the world will see marked progress on the problem of not just whole brain emulation, but also transfer of human intelligence to new substrates. That's very exciting. This is a new sci-fi corner. I love it. Every sci-fi trope everywhere all at once. Time travel?

Is time travel on the table? We don't know yet. So it's an interesting question. We don't know whether the physics of our universe are compatible with, I think what most folks would construe as time travel. There are versions of temporal non-locality that are consistent with the physics that we have right now, but not in a useful way that would be worth filming a sci-fi movie over.

What about faster than light travel? We don't know yet. It's the same problem. In some sense, exactly the same problem as time travel. We don't know whether the physics of our universe will ultimately allow that. Because it seems like if you don't get FTF, it's going to be a boring. travel to go to another solar system. I will say this. I think odds are pretty good that AI in the

next few years will tell us whether the laws of physics of our universe are compatible with that or not and help us solve this. One of the reasons why I helped found physical superintelligence to discover any of such laws. Very cool. Well, thank you so much for taking the time to come chat with us. Yeah, great to meet you. Congratulations on the progress. Fascinating stuff. Thank you. Lachwest. We'll talk to you soon. Cheers. Have a good one. Let me tell you about Figma. No matter where are your Starts, Figma Make, quad code codex or a sketch. The Figma canvas is where ideas connect and products take shape, build in the right direction with Figma. And let me also tell you about Cisco.

Critical infrastructure for the AI era. Unlocked seamless real-time experiences a new value with Cisco. And without further ado, we have Charles Lamanagh from Microsoft in the restroom room. Now he's in the TV available from Charles. How are you doing? Good. Thanks for having me on, guys. Thanks so much for coming on the show. Why don't you give us a little bit of an introduction on yourself and then we can get into the news? Sure thing. I'm on the product and engineering side at Microsoft. My team builds out things like agents, a bunch of the platform work and our business applications and some low-code stuff. So a bunch of odds and ends. And what's the latest and greatest in your world? Yes, a lot of big news today. So today we announced co-pilot, co-work, which really takes co-pilot beyond chat. So it can take action.

and automate tasks and pick delegation, basically. We had some new agentic capabilities in the office apps and a bunch of agent management solutions. Yeah, how are you thinking about the open-claw boom people writing code locally versus Microsoft has Azure? Like, it's so easy to throw everything in the cloud. And I feel like one of the barriers that a lot of people have is like, do I need to go get a Mac Mini?

Can I just delegate to the cloud? some of the apps don't just by default build an app in the cloud or they don't have functionality. How are you feeling about that tradeoff these days? Yeah, I mean, we're super excited by OpenClaw and all that means. I think it's like the resurgence of the PC. Like what's the best thing you can get in a computer? Let it run the apps. Let us log in as you.

Let it messages you. I think like a Windows PC in the cloud is probably the main way people will ultimately use something like OpenClaught at work. Yeah. How do you think about model selection? Obviously, Microsoft has a time-honored partnership with OpenAI at this point. But, you know, Microsoft has their own AI lab. And you can get pretty much any model on Azure. Do customers want to pick models?

How do you think about what is the best tool for the job in the products that you're building? But then also, how much do you want to give to the user? Because there's sometimes when I go into a chat app and I'm like, I just want auto. you should be figuring this out. I know the details of pro versus thinking, but I don't want to think at all. Yeah. So that's one of the things that we think is super important

for co-pilot and a lot of our products, the fact that it's multi-model. And the reason is because the best model for a given kind of task is not always the same, and it's not even the same from month to month or quarter to quarter. Like, if I look at Microsoft 365 co-pilot,

chat is like the main way people interact with it. Our auto router is going to put you in open AI models because that's what we think works best with our work grounding in our work context. But then if you go to co-work, the new announcement today, and you fire off a task and you want it to run for a long period of time in the background, use a bunch of tools, maybe generate some code, use like the terminal and stuff. That uses the clot family of models. I go to co-pilot one place and get both,

and then we're going to keep adding more models, models like from MAI and other places where it makes sense. But it's really about the experience the user wants, not so much picking exactly the version or the type. Yeah. How have conversations been with business leaders in terms of deploying a product like co-pilot co-work? Because the ramp up for software engineers was smooth and everyone was aware of it, tracking everything, and going from better auto-complete to agentic coding was pretty intelligible at every step of the way for most software developers. but for someone who's lived their life in spreadsheets and email and maybe not really thought about building automations all that often, what does actual deployment of these tools look like?

Yeah, it's such a great question because I think there's like two pieces to it. The first is you have IT teams, which are nervous with, if you imagine every single employee spending up a whole bunch of agents that can create documents, send emails, run workflows. So they want to make sure they have the right governance and controls in place. That's why another big part of the announcement today was something called Agent 365, which makes it easy for IT teams to manage and deploy agents from Microsoft and other companies. So that's one part. And then you have the end users. How do you teach a typical Excel or PowerPoint or Outlook user this idea of agentic working?

Even in coders, it's taken time and it's probably not fully diffuse. So that's the thing that we think we're in a really unique position at Mike. unique position at Microsoft in terms of being able to do because we can bring things like co-work to users where they already spend their time inside of the office apps, inside of Windows, and increasingly inside the co-pilot app. And we can present it in a way where it's digestible to hundreds of millions of office users, not just tens of millions of coders. Is the majority of your business model still seat-based at this point? Because I imagine that there will be a transition at some point, how much of that transition is already happening?

Yeah, we're still very seat based on like the M365 side, but like the good news that Microsoft, we kind of know how to do consumption businesses because we got Azure. We got a bunch of other very large capacity businesses. So the fact that we have user licensing and then consumption and pay as you go licensing for capacity all integrate with one commercial kind of vehicle with customers, we think we're in a pretty good position to make that transition over the next couple of years. There was some chatter on X this weekend over how different labs are pricing tokens and plans, potentially to take market share from each other.

There's a little bit of a capital fight playing out. How are you communicating? They are more specifically competing with the application layer. Yes. And so my question is, how are you talking to business leaders, CTOs about how they should think about forecasting token budgets and spend and compute budgets. If they're used to, okay, I have licenses and licenses maybe go up in price and my head count goes up and I'm used to that.

But all of a sudden it's like, well, everyone's paying $200 a month, but it might be $5,000 a month in a couple of months. But also the models are getting cheaper and the inference is getting cheaper and the capabilities are going up. And it's very hard to predict. What are you talking to, what are you telling to business leaders and CTOs? Yeah. Yeah.

So the first thing is you have to embrace this technology, otherwise you're going to be left behind right now because the productivity gains are just so significant. And if you go talk to a coder and you told them, hey, I'm going to take away like GitHub co-pilot and the agentic coding capabilities, they'd be like, I refuse to work in this environment. Like it's just inhumane almost. So I think the same type of thing is going to happen for all information work, all office work. So nine months from now, I think if you went to somebody and said, we're going to take away your agentic tools like co-pilot cowork, they'd be like, no way, I'm not going to go back to the old way of working. So I think the first is that there's a degree of inevitability

because the benefit is so large, and there's such strong pull from the end users. That said, everybody has budgets. So what you're going to see and what you're already starting to see is a bunch of controls where IT teams can start to manage exactly how many tokens or how much dollars each user can spend and they're going to allocate that in a way based on where they think they're getting return and diminishing return and it'll probably be like a expense report approval hierarchy type thing so you'll see all of that show up but part of just doing

work now is you need a token budget that that's the future of how we think about in how you budget for a team or a department where do you have any exciting case studies on the non-sophobic engineering, agentic work side of the business. Because when agentic coding came to software development, you know, there was a shift in a lot of software developers' workflows where they went from writing a lot of code to reading a lot of code, reviewing PRs. They started operating at higher and higher levels of abstraction,

but the decision what to build, how to prioritize things, how to build features, there's still conversations and software engineers still doing work. What does that look like in the context of a finance department or operations department or a sales group? Like where, when the rubber meets the road, like, where are you seeing? Like, okay, this, this is ready. This is the next sub area or category that will really, really benefit from the productivity gains from agentic workflows. Yeah, I mean, a general thing, and then maybe a more specific thing. What we're seeing is, like, how do you measure productivity?

Is itself a challenge? I mean, this predates AI, right? Like lines of code, number of pull requests, number of bugs fixed. I don't think I've ever heard a good engineering leader say, that's how I measure the success of my team, right? It shows up in other ways. Do you have a great product people love? Is your cost structure good?

What's your install base look like? So I think we have to get better at capturing that final business outcome because that's what's going to get the CFO to open up the pocketbook and continue to invest in more tokens and more AI. So if I go then take that and apply to kind of other disciplines, what you don't want to say is, oh, our salespeople send a lot more emails or our finance professionals create a hundred more spreadsheets.

That's more like of a nightmare scenario for AI. One billion spreadsheets per company. No, no. Exactly. Yeah, we saw someone on the timeline say that their token budget went way up, but their revenue didn't go up. And so they're like, I don't know if it's, if it's worth it. Like you got to show results. Exactly. So that is what the question is going to be

everywhere. But the good news is engineers historically have resisted clear measurement of impact because, you know, there's a little bit of part to it. But like sales professionals or finance professionals or customer service and support reps, they are so metric driven. Like if I'm a sales rep, what's my quota? What's my attainment? What's my revenue per head? Every single sales department has all of those things instrumented. And we've seen with M365 copilot, we kind of did a bunch of experiments and we published a white paper on it, like an almost 10% increase in revenue per seller just by giving them the family of M365 copilot tools and things like our sales extension. Like that's real dollars in the bank. Same thing for like back office finance processes. We've

seen people use like our agents to do like accounts reconciliation. And we have like a great case study we did there, whereas like 30% of all back office finance was automated out by this thing. So those are examples of real top line and bottom line benefit. And that's what everybody's going to be talking about in the next couple of years. It's not going to be, you know, emails, documents, files, etc. Yeah, yeah. I mean, you're in a, you're in a high level business role yourself. I find myself doing a lot of like knowledge retrieval, deep research reports, but have you personally been using a gentic, I mean this product specifically in your day to day?

Like, even as something as simple as like a, Sam Allman told us that he has a self, he has some sort of like to-do list that does itself or something. I still don't understand exactly what that product is, but how are you using these tools these days? Yeah. Yeah. So the, for like copilot cowork, for example, I've loved that. Like my favorite magic moment has been around calendar management. Oh.

I was telling the story to sell. one like I go look out three months in the future and try to see you know how crazy is my calendar and travel look like interesting and I was able to decline 17 meetings for me okay that's right there yes exactly nothing gives me more joy than declining a meeting you sold me yes and I say like I went the outlook and just saw all the meetings disappear and it was like such a serene moment for me and and like it did a deep analysis yeah it went and said like who's my manager who do I work with Am I optional or required? What's the topic?

And it recommended the 17. And I went through all 17 and said, you know what? Decline. And it was right. So, like, that's one thing that I use it for. Like, another thing that I like right now is, I mean, I'm an engineering background, so I'm not going to go create the most beautiful content all the time. So I use it now to pull together information, design docs, media notes, and then go create

either like architecture diagrams, presentations, or spreadsheets or docs. And, I mean, I know it's just more dock generation. but it's very much more kind of like fire and forget. So it works really well. Yeah, yeah. I mean, even if it's just like the bones of a PowerPoint presentation with the actual data pre-populated that would just previously be a lot of copy-paste,

go run this analysis, copy it over, make sure it's the same font. Like there's so much grunt work that can be automated with these tools. I'm super excited for that. Where do you see the design PowerPoint generation tool use? I mean, people have been talking about, like it's hard to create an RL environment with a verifiable reward around a great PDF, but it's easy to do that with coding. Do you have like an internal data set of like the best PowerPoints ever created in your that you're like RLing against?

Or where do you see that going? Yeah. I mean, I think that is something that there's a little bit of taste in all of this because like depending on the company, some companies like lots of infographics and lots of colors. Yep. Other companies like super fact based, no image. So, like, you know, there's a little bit of an eye of the beholder, but absolutely, we have a very robust harness and, like, gym environment where we're able to quickly evaluate models,

prompts, tools in our own, like, fine-tuning efforts to see if it's producing better content. So I think, like this, like PowerPoint's, word documents, spreadsheets, it's going to be like code. And you're starting to see it already, but it's going to be like code for sure. Yeah. I mean, like, a lot of these things. or like XML under the hood anyway and can be puppeteered like code anyway. Well, congratulations on the launch. I hope you don't mind.

I've been calling co-pilot co-work Coco. Coco. So hopefully it doesn't stick unless you like it. It's very cute. Yeah, it's a very cute name. But congrats on the launch and great to meet you. Thank you so much for taking the time to come on the show. Cheers, Charles.

We'll talk to you soon. Let me tell you about MongoDB. What's the only thing faster than the AI market, your business, on MongoDB. Don't just build AI. Own the data platform that powers it. And let me also tell you about graphite. Code review for the age of AI. Graphite helps teams on GitHub ship higher quality software faster. There's an interesting data point from Apollo featured by restructuring here. AI is making, starting a business easier and the numbers confirm it. Do you see this? Weekly business formation is exploding higher. So, previously... This matches what the Carlson brothers said

when they're seeing not just more business formation, but businesses growing faster. It could be that AI is actually making businesses harder to start, but Stripe Atlas is making it so much easier that we're seeing an upward trend, and it's fighting back against AI. No, of course. These are...

These are multiplicative very clearly. But Stripe Atlas does make it very easy to start a business. US business applications jumped insanely after the COVID pandemic, in 2019, we were seeing about 60,000, 70,000 new business applications, I think a four-week moving average. We've been over 120K for years now. And AI seems to be driving that trend upwards by about 20% over the last two years, which is great news. More small businesses, they are extremely edifying, extremely valuable. I highly recommend. We got another white pill.

Give it to me. The data center buildout has allowed Ludin County to lower property taxes by 38% since 2010, translating to average savings of about $3,400 per year for homeowners. This is great. Love it. What's happening here? Data Center square footage in the millions has gone from 5 million square feet to 40 million square feet. That's a huge jump.

Yes, this is like Northern Virginia. Yeah, this is like AWS East one. Okay. Okay. Yeah. So homeowners are paying. They went from 1.2% of whatever their house is. So if they have a million dollar house, they will be paying $12,000 a year. And now they're at 0.8% property tax rate because they're taxing those data centers.

This is the easiest solution to every AI problem. If the companies are profitable, tax them. If we already tax every business, we tax them. We tax them by default. If somebody makes a lot of money on an AI stock, taxes its capital gains. You get the money. The money shall flow. And so good news for the folks over in Moondon County.

Jacob says America once named things plainly, place trade surname. You had international business machines, long distance telephone. Studebaker. Tudabaker. And Bethlehem Steel. That international business. machine's logo is amazing. Is this real? Or is this like a reimagining of what the IBM logo could have been? Because I love this logo. What a very cool thing. That is apparently the original IBM logo.

Wow. We don't know how to make logos like that in this country anymore. Seriously. I mean, that's what kicked all this off. Somebody was upset about the granola rebark. America's innovation slowdown is primarily due to the amount of Lord of the Rings. that founders have to flip through before identifying the name they can raise off of. No, you can just start a company and name it international business machines. Whatever you want to do, just name the company exactly that. Technology and business programming network. It's that easy.

In other news, Two Chains. Two Chains revealed he was an early investor in SpaceX. This is huge. He says, it just felt like I got in like early Bitcoin. There are some crazy early SpaceX investors. I don't know what Elon was doing with that early road show. We know some of these folks. Very cool community early SpaceX.

Get me two chains. Two chains got in at some point. And Gabriel over at OpenAI says two chains is about to be three chains. Well, let me tell you about OCTA. Octa helps you assign every AI agent a trusted identity so you get the power of AI without the risk. Secure every agent, secure any agent. Let me also tell you about the New York Stock Exchange. Hopefully we'll be seeing SpaceX there soon.

Want to change the world? Raise capital at the New York Stock Exchange. Do you imagine two chains and Elon ringing the bell at the New York Stock Exchange? I can. I hope it happens. Well, we have our next guest in the stream reader. Julian back from Sequoia Capital. His partner there.

Welcome to the show. Julian. Good to meet you. Hey, guys. How you doing? Nice to see you. We also hope to ring the bell very soon. Holy logo mug.

Yeah. Holy logo. I know. I know. I know. I know. We're partnered with this. Thank you for backing them early, making TVPN possible.

Through your... 30 years ago. You paid the way. Since this is the first time of the show, why do you give us just a brief back story on yourself, how you wound up as the Sequoia, what you focus on, what your day-to-day is like? Sure. So first, thank you both for having me. It's an honor to be there.

I know you had a lot of my partners as well, so we're very excited to be on the show. My name is Julian. I'm a partner on the team. I'm based in London, but it's been a lot of time going back and forth to spend time with the rest of the team in California. And I started as a founder a very, very long time ago in high school, built companies in hardware before. And so that's why I've been investing mostly in software for most of my career. Wow, you were like this stuff, I got to get out. I was like, man, it's hard.

Are you thinking about going back to hardware? Is software dead? What are you thinking? Yeah, well, I mean, two things can be true at once. I think hardware is really interesting. I'm spending a lot of time thinking about physical AI. But at the same time, I've been investing for the last 10 years in application software. So I want to remain true to that.

And this is kind of the things I've been spending time on recently. Yeah, take me through your current thesis on how software is changing in the age of AI, in the age of, you know, maybe some sort of takeoff. AGI is here, according to Sequoia Capital, amongst many other people. The labs are recursively self-improving. And it seems like, you know, Sam Altman made the claim that, like, if you want to build a startup, you shouldn't build a company that is in the direct path, while more and more products seem to be in the path of AI and AGI.

How are you thinking about the changing dynamic with software as we know it? Yeah, that's a really important question. And I think we've been listening to Sam a lot as shareholders. And recently, just talking to our founders, a lot of them are wondering if they're just an iteration away from the models replacing what they're doing. And it puts us in the position of thinking what is coming next, what is going to be defensible. And for us, we've been making that prediction last week that the next trillion-dollar company will be a software business that masquerades as a services firm. And we share that thesis with everyone online.

The reality is if you sell tools today, you're really in the line of sight for the models. And you're effectively competing with the next generation that they're going to launch, Whereas if you sell the work, you're actually benefiting from what the models are doing and all the billions of dollars that are going towards AI. So part of that prediction was saying for every dollar that's spent on software, $6 are spent on services. Until now, we could only really go after the $1 because we were building tools. Now that we can actually deliver outcomes, companies should think about capturing those $6 instead.

So the, I mean, the obvious, like, first step is just get to outcome-based pricing, move away from a seat-based model, probably. Would you agree with that as sort of, like, the most immediate change in business structure that needs to happen? Or if you're starting a new company, maybe don't start with seat-based pricing on day one, or is the, like, seat-based pricing is dead? Maybe that's, like, an overblown characterization. How do you think about just, like, the business?

model moving forward? Yeah, we think both can work. They just serve different purposes. We have companies in our portfolio that start as these co-pilots that are helping the workers. And then we call them autopilots, the one who, you know, do everything into end. Many of them are transitioning from co-pilot to autopilot to do what you're describing. A company that comes to mind is a business called Sierra that started by Brett Taylor

that basically helps companies with customer support, which is pretty direct bullseye use case of AI. And in this example, they're actually charging the customers per outcome, right, per resolved ticket. And so if you spend $20 per ticket resolved with humans, they charge you, say, $5 instead, right? And that's driving ROI for the company. It's driving an outcome that's very measurable.

And so that's the first more obvious. company that's really working doing that. Frankly, until recently, outside of software engineering, we haven't seen a lot of this sort of, you know, fully own end-to-end outcome driving.

But the models are really progressing quickly. And for everyone who's been vibe coding the last couple of months, I think we've all been impressed that we can code a full websites, just end-to-end, without any skills, frankly. And basically our prediction is

that this is going to happen to other categories. And we've been mapping out these categories and helping founders come up with frameworks to prioritize where to attack them and how to do it. Sorry to interrupt. One thing I'd be curious to get your take on. So if a company is just selling enterprise software that helps people accomplish a task, that feels very threatened by a foundation model company that maybe makes an agent that can also do said task and then you don't even need the software.

So if a company transitions to just selling the work, selling the service, like you're saying, you would agree that they're still going to be under threat from the labs, who will also presumably be selling intelligence or agents that can go and do the work, but you're saying it will maybe be more closer to a fair fight. Is that the right way to think of it? because I still think, let's say if somebody is like spinning up a, instead of making a legal AI tool, they make an AI native law firm and they're selling the work, I could still go to a chat Chhabit, a claw, a Gemini, and a future version of the model, I can just say like,

hey, I need to do this thing, can you help me do it? So there's still going to be that competition, but you're saying that's a much better sort of competitive environment to be in than just selling like the software that enables somebody to do the task. Is that, is that roughly right? Or is there something I'm missing? Yeah, I'll just add one framework that I think is helpful for people to clarify what, what that means. Basically, everyone is conflating the model's intelligence with what human are really good at, which is judgment. And I think if you strip those two concepts that you really understand what the future will look like. And so right now, the model are becoming really good at everything intelligence-related.

So think of tasks that are rules-based where just logic gets you to the goal, right? Whereas the judgment is everything the humans really spike at, which is more harder to define, right? It's instinct. People call it taste, call it experience. And those things are things we think the humans will remain really good at for a very long time

until intelligence is able to absorb these skills into it. But it's just really hard, right? Because we give the models a lot of credit for their intelligence, but they're just trained on set amounts of data. And so we actually think that this hybrid motion of coupling the intelligence from the model with the judgment from the humans is going to help you build the next generation of companies,

which we call these autopilots, right? such a true example, in the law firm, you will automate a lot of the intelligence work that's very much codified, which is what the model will do. If you put the employee at the center, someone who's really experienced, who can talk to a jury, who can, you know, mark a pause when at the right time, those are all the things that you learn only from real world experience that a top lawyer will be able to experiment, but that the models might will omit today. Does that make sense? Yeah, I still, I still, like, you know, maybe, maybe I'm, maybe I'm too AGI-I-pilled, but I just assume that a sufficiently advanced agent would be able to capture and accumulate, you know, the ability to. Two-AGI-I-pilled.

Just talked to Dr. Alex Wistner Gross for 30 minutes, and he's like, I'm uploading my brain right now. No, it's not that, but it's just the idea of like a, you know, you put a smart person, into an organization where they don't have any experience and they start doing things and they start getting feedback from other humans in the organization and over time they accumulate enough knowledge and domain expertise that they're able to excel at something that even using the customer service example you know you can you can you've always been able to outsource customer service and you can go work with a firm that is excellent at customer service the alternative has always been to effectively hire a really great leader for that part of the organization, have them build out a team

and a structure and an approach, and execute something that would be a similar result. There's like a cost trade-off, but... It does feel like overall there's like maybe some pressure on the B2B2B model or like the B2B2C model. if instead of, to use your example of customer support, there was a time when software companies would sell phone tree software to, not to companies directly to do customer support, but to customer support outsourcing agencies that would then deploy that. And now you're seeing more companies go directly to, like, the firm that makes the customer

support experience end-to-end, go directly to the company that needs customer support, as opposed to having this, like, middle level. And it almost makes me feel like we're maybe going to rotate even a little bit further into D to C. When I think about law, there's a lot of stuff where the software would go to the lawyer, the law firm goes to the person, and if the person can go straight to the law firm or the legal model, you wind up with a little bit more B to C activity, a little bit more company directly to the service provider or the technology provider. But again, this feels a little bit dependent on specific market dynamics. I'm interested to know your reaction to that, but also what does hold in your mind in the near-term AGI scenario? Because there's been a lot of debate over

how will marketplaces perform or are network effects still super sticky in a world of AGI? Or what are the powers, the seven powers? Yeah, I guess part of what I just kind of close out my thoughts. Like I agree, I agree with the entire thesis in like the relative short term, like the one to two years. But then beyond that, there's still some uncertainty around. There always is, yeah. Like, I don't think it, I don't think it solves the sort of discount that even traditional SaaS companies are getting today. Sure, sure, sure.

Sure. Yeah. On that point, I think you're right. And basically what we realize is that no one really knows, right? I've been in many partner meetings recently where everyone just is astonished at how quickly things are moving. I've been doing that for 10 years and it's just by far the fastest rate of change I've noticed. But some of my partners, you know, Alfred Lynn, Pat Grady, they've been doing that for a lot longer. and they also acknowledge that everything is moving so quickly.

And so what we're trying to figure out is also not what is going to change, what are the things that will remain true no matter what, right? Customers will always want things faster, better, cheaper. And so that's kind of the thesis that we're describing. There are just categories that require more intelligence than judgment. And so we think that in the immediate, the ones that require more intelligence, will be actually really interesting for these autopilots because you can have,

instead of having 10 humans in one AI, you have one human in 10 AIs, right? And just that ratio is just going to shift as the models get better and are able to absorb what we use to call judgment from the humans that is turning to intelligence. And just to put like a practical framework to that, you mentioned outsourcing.

It's actually the best wedge right now is just to focus on outsourcing if you want to capture this services span because there's three reasons. The first one is that outsourcing has already been agreed to a third party. So it's very easy to sell, right? Because it's already a vendor

that's external to the company. The second is that there's an existing budget that's assigned to it, which you can pay yourself against. And the third is that it's actually very straightforward to swap a supplier. It's not so straightforward to reorganize. your company, right? We all talk about headcount reduction, you know, in terms of ROI,

but actually putting that in practice takes time, whereas swapping a vendor is very quick. So I actually think that, and that's the prediction, is that the companies that will succeed very well with this autopilot model will start with where there's a lot of labor spend that's being outsourced today. We mentioned customer support. We'll mention also things like accounting, tax, insurance, and so on, so forth. Is there a world where in the next five years or something, like venture capitalists truly shift away from software and go back into rare metal mining or something crazy like that? I just keep going back to like Daniel Gross, now it met a venture capitalist. I think one of his first AGI trades, the first bet was like in the world of AGI is copper mispriced.

And it was. It doubled in price over two years. And I'm just reflecting on this idea that like if the Java venture capitalist is to back software companies, that's one thing. But if it's just to allocate capital in a way that can have exponential upside and single investment to return of fun. Would it ever make sense to stop investing in software? I would hate that because we obviously love enterprise software. But it is been funny. A lot of people have been pushing like, you know, the model. models are so good, we just need better harnesses.

Yeah. And to me, this is giving me flashback to early GPT days where people were like, GPT is really powerful. You just need the right harness. Right. All these like wrappers. Yeah, yeah. And had explosive growth very momentarily until, and so I do.

Most AGI-I-pilled person ever over here. I love it. But what do you think about, about like, the more physical aspects of the job. Rare metal mining, for example. Does that seem like something people will be working on a couple of years? Yeah, I mean, everyone's talking about the fact that energy will be the biggest bottleneck, right, to train these models.

So it's very natural that you would go there. Personally, that's not really where my skill set is. So I really hope you're wrong in that we'll find other areas in between. I think the real path will be similar. in the middle. Hopefully, right? My way of doing things has always been to focus on talent. And so right now I'm just always asking the smartest people I know. Who do you know who's the smartest person? What are they working on? And you're right. More and more people are going more into the physical world. But right now that probably means things like robotics and

things like that. People are making a lot of money trading commodities, but that's a different world. Yeah. Who knows? Maybe that's the job of the venture capitalist in 2030, just trading copper all day long. You mentioned using AI to build a website despite maybe skills that have languished since your time in software or hardware, but how are you using AI tools these days? Are you still doing mostly like knowledge retrieval, market research for your job in venture capital? Or have you actually built tools that have accelerated certain things? Have you automated any of your workflows? I'm just kind of curious how

people are using AI these days. Yeah, well, I don't know. I hope my competitors are not listening, so they are. So you can say, I haven't touched it actually. It's useless. Yeah, exactly. Exactly. You should

uninstall. Stick to organic capital. Pen and paper, actually, is underrated. Exactly. Back to the basics pen and paper. and really don't try cloud code. It's really, really great. No, jokes aside, it's been really a transformation

and particularly the first quarter of this year. I mean, I use it for all kinds. You know, the work is basically sourcing, picking, winning, company building, and harvesting, right? I really have workflows that fit all these boxes. So it goes from things as, you know, at the end of the week, it will just like scan my entire calendar, look at all the meetings I took, look at the granola notes, which is a transcript,

do an analysis and help me, you know, try and find the key questions and the due diligence. And, you know, it'll help me just prioritize, right? Because this job is just so much about prioritizing. It's a power law, right? There's only three meetings in the year that will really, that could change the course of your career. And so making sure that you're attuned in those meetings is really important. So that's really what I would say Claude has helped me do right now. And I'm sure, you know,

opening I will release something next month that will be amazing and so on. You just have to experiment, right? And I think that's also like what we've seen with things like OpenClaught being so top of mind is like this culture of experimentation is not just in software engineering now, but it's transpired across the entire knowledge workforce. And, you know, venture capitalists are not an exception. I love it. Well, thank you so much for taking the time to come shout with us. Yeah, great to meet you. Have a great rest of your day. Thanks for doing that. We'll talk

to more. Thanks for doing that. Great time. Let me tell you about 11 labs. Build intelligent, real-time conversational agents. Reimagined human technology interaction with 11 labs. You know, if I was a venture capitalist, open claw. Every morning at 5 a.m., text me the name and market cap of the company that I passed on three years ago. I need inspiration to grind harder. That's the secret to success. As is Lambda. Lambda is the Super Intelligence Cloud, building AI Supercomputers for training an inference that scale from one GPU to hundreds of thousands. And without further ado, we have Owen McCabe, the CEO, the CEO, a huge supporter of TVPN as well. Welcome to the show. How are you doing? Great to see you. Great to see you. Great to see you both also. It's been too long. It's been too long. Man, you have, you just have a voice for for radio. Live streams, podcasts, all of the time. It's incredibly calming. Yes, I love it. Okay, what would you like me to say? I could record a meditation for us. We could kind of start at the hour that way. You have time? Do you have time to meditate? Do you have time to meditate these days? It's kind of the only thing that keeps me

sane. That's good. Well, you have some new funding for new project. Break us down. What happened? What's the plan? How much did you raise? Take us through it. Yes, we raised $250 million. Let me hit the gong. Congratulations. Hit it. Did that gong get bigger since I was here? It did get bigger. It did get bigger. It will be cool if it got bigger every single season, right? Fast takeoff.

Just wait. There's a fast takeoff. There's a fast takeoff in gong size over here. It's growing exponentially. Except this was actually, I think, the largest we could find in the U.S. So we've got to go. There's always bigger. I kind of have gong envy right now.

I wish I had a gun. We'll send one over. Let's talk about that later. We'll talk. We'll talk. So $250 million, what's the plan? Yeah, yeah. So, I mean, we've been building this thing called Finn for three years now.

Service was just this obvious place where AI was going to, you know, blow up. And Austin and everyone else started to chase it. We were about a year ahead, everyone else. And we started doing, you know, customer queries and then more complex problems. Originally it was answering questions for customers, and it's fantastic, you know. The typical response rate with a human was one. two days, you'd have to deal with the beautifully perfections of humans. Unfortunately,

these bots are just far more perfect than humans, and they answer in seconds. And of course, they're cheaper, too. So we started to say, Better, Foster, cheaper. These things are really better, faster, cheaper. So that was a tremendous success. Finn is the biggest service agent in our category, nearly 100 million error or 8,000 paying customers.

You know, some of the biggest lows in the game. Thank you, from Anthropic to snowflake, the cool guys like Bollie Market and Vent, and everyone else are on Finn. So, wonderful. But where does it go next? Clearly, it's just so obvious as each of these categories developed

that there's just more to explore. None of these categories are in any way fully baked. And that's one of the interesting things about kind of all these AI spaces that we're still kind of pulling the thread and seeing what's going to show up next. And the next big play for Finn is not just answering service queries, but really getting deep into every stage in the customer life cycle. So we call that the customer agent, and this money will primarily bring the customer agent to market. Okay, so when I want great customer service for a company that I'm running,

I want, yeah, you know, search the knowledge base, answer the obvious questions, handle the refunds, track. That's part of the customer journey. Are there other touchpoints that are unlocked by AI along that journey? Absolutely, completely, yes. from day zero on the website, when the customer has very basic questions, when they get more interesting, when the business wants to start to qualify and understand who that customer is, when the customer actually then wants their hand-held, when they sign up,

or they want to actually engage and learn and meet salespeople, they want to book a demo, or they want to talk to a virtual salesperson. I mean, that's just the very top of the funnel alone. But to get into the application, if it's B2B, there's all the onboarding, all the detailed help required to get something successful. If it's e-commerce, it's actually buying products and getting it into the cart and bringing them back for more shopping. So there is really the literally the full customer lifecycle of work to be done by agents. And customer service was just the little sliver.

I want to know what's happening off-camera, though. Oh, we just have our next guest walking into the studio. Got it, got it. So I'm just... Hopefully that's interesting to me. Yes. I want to talk about some of the best customer service. agents that I've hired have sort of wound up in...

Real people. Real people have wound up in sales roles almost. Like, they have a fish on the line. They're reeling them in, saying, you don't want to cancel. We actually, this product does exactly what you want. And the fact that it was delivered one day late, we're happy to refund you.

But it seems like the underlying issue is that you actually want this product. And so this product, why don't we get you over here? And that's this like sort of different. skill set that comes out in the best customer service agents. How are you thinking about actually driving incremental sales through customer service agents? Yeah, I think one of the interesting things is that a lot of the work in these functions isn't entirely separate. We just needed to draw lines between them when we were running human organizations because

you can't possibly have one person who's outstanding at service, sales, marketing success, etc. Yeah, and the SaaS market, it would make the SaaS market maps too complicated. Totally. A pen in the ass. But unfortunately that pen of the ass is coming with AI because I just don't believe you're going to have these separate sales agents and marketing agents. You're not going to have these things fighting with each other, having different contexts and information, different memory and different parts of the customer life cycle, different messengers, not going to happen. There's going to be vendors that will provide one seamless solution across the customer life cycle. And that's a big thing we're passionate about. And one of the, very fascinating things we've seen. We've been building this for quite a number of months now, and we're going to launch the first part of it next month. One of the fascinating things we've seen is that when there's an underlying relationship with the agent, and then there's actual connection in the moment, and then there's clear intent, the context is primed for selling and encouraging the user to do something next. If you're on a website and you get a super

annoying pop-up that says, hey John or Jordy. In fact, they're not even going to use your name. They're going to say, do you want to learn more about or whatever? And you're like, get out of here. It's like not the right time or place. Whereas if you're actually discussing, it can ask you questions back. And it's so much more effective. And so we've seen just these things.

The service agent, when it starts to do sales, for example, and vice versa, they're far more effective. And so the real magic from AI across the board and, of course, this agent, world is going to be when the agents can do things that humans can. Yeah, one thing that stands out is like right now you have this spikiness and dialogue as a customer of a business, right? You might have a question coming in and it like creates a moment and then it goes away. And then you get maybe transition to someone else and you're talking to a real human.

You talked with them a little bit and you get past to someone else. And like there's something about creating, I think the customer experience of having this continuous conversation with the organization that you're, you're a customer of that I think will be super powerful because it just means that, you know, a split second away the entire time. I was talking with some guys that own a number of like big car dealerships recently. And I was talking to them about AI and they weren't fully sold yet because they use like chat GPT, but but they haven't had anything spread across their whole organization. And I was explaining to them how there's so many different moments where like a

customer lands on their website and they want to know if a car is available, but somebody gets back to them too slowly. And like the light was like finally going off for them. Do you think that there's a number, how many industries out there have been kind of resistant to even using software from Silicon Valley to date that are going to kind of take a second look at all the different tools and things like Finn now that they maybe have had a cool experience with AI? in their personal life, and they're basically saying they're willing to kind of look because maybe the historical SaaS platforms, they tried them, but it wasn't quite magical yet. Yeah, there'll be so many spaces, so many verticals that some smart entrepreneurs will benefit

from bringing sexy technology to that aren't immediately obvious to those of us in Silicon Valley. You know, we tend to sell to ourselves. It's really insular. So all the names I just mentioned there, Anthropic and Snowflake. They're all the cool brands in our world. However, there's like car dealerships that are, you know, I don't know, maybe it's not as Anthropic, but they're pretty down big. Depending on the car dealership,

I would put car dealership above Anthropic and Snowflake personally. If it's like a Ferrari dealership, Lamborghini dealership, like that. Just on the cool side, we got to give it to a super car. On the aura factor, sure. For sure. Maybe not on the economic factor. Yeah.

But there are so many industries that just don't have our attention. And so there will be, you know, there'll be companies like ours sell into many verticals, but really, you know, we'll focus on all of the verticals that are adjacent to ours. And there's, you know,

tens of billions of dollars opportunity just in that alone. But there'll be people who will say, we're going to make agents for dentists, car dealerships, for certain government bodies, maybe even for tax bodies. Like, there'll be people who do this

because so much of the effort is not just in building the technology. It's actually creating the trust, as you say, with the actual people who will have to deploy this technology and make it work for them with their legacy systems. So there's a lot of very unsexy work that will be richer rewarding for some people who bring into these places.

Talk about the decision to raise debt instead of equity, not something you see. Yeah, I mean, yeah, I know. I just realized I had this kind of blind spot for this as an option for capital. And there's just so much equity opportunities out there. We raised a bunch last year for a secondary. But when you're in our dynamic where you're accelerating as quickly as we are, like every year we're doubling our rate of growth pretty much, there's no way we're going to be able to achieve the valuation

that we're pretty confident we'll get next year. And so any equity raise is going to just be massively dilutive. We calculated that raising this in debt is going to cost us about a, 10th, the kind of price of shareholders that it would cost in equity. So this is, you know, the incumbents have disadvantages. They tend to move a little slower. They have older businesses they need to wrangle with. But they have access to a phenomenal amount of debt.

Like 250 million is a relatively small amount of debt for the size of our business. And I think that you're going to start to see a lot of these older SaaS companies as they pivot to AI and they actually find new growth, start to lean on. on debt as an opportunity to invest rather than, you know, super delude of equity. And honestly, I just think it takes a little bit of maturity to take that route. There's so much pressure from employees. It's not just maturity. It's confidence. Yeah. To know your business.

Well, you know, I like to think so. But the reason I say maturity is that there's so much temptation to go and attain that latest, honest valuation. Your employees want it. Your previous investors want it. media wants it. But if you're able to tell them we don't need it, actually it's way cheaper for us to get a lot more capital with far less effort, then you can actually spread the needle in a way that I think is going to work much better for big companies. The gong over here rings just as loud for debt as it does for equity. So don't let the media. Yeah, I didn't actually hear it. You guys are hiring a ton. You said you're bringing on 650 new people this year.

Yes. Yeah. So that's gross. I think net it'll be a little less than that. It's a phenomenal amount of hiring. I fear it. I celebrate it with great caution. There are just countless stories of the hubris of CEOs who've boasted about how many people they've hired and it's all ended in tears.

So it's actually not a point of pride, frankly. I think small is beautiful, but we are growing incredibly fast and we need all the help we can get. And you've been at this long enough to know, the impact of what 100 people, 600 people, what impact that will have on your organization. So I have the full faith in you.

Thank you for that. Thank you so much for taking the time to come chat with us. Great to see you. Congratulations. Good to see you both. We'll talk to you. We'll talk to you. Thank you.

Have you. Have you done. of scaling, monitoring, and security. And let me also tell you about Century. Century shows developers what's broken and helps them fix it fast. That's why 150,000 organizations use it to keep their apps working. And without further ado, we have Alex Epstein. He's the author of Fossil Future live in the TVPN Ultridum.

Welcome to the show. Good to see you. Thank you so much for coming down on short notice. What are your, what are your text been like the last 48 hours? This is over. Because John and I were thinking, who's our oil guy? We had one. We had one. And I'm local too.

Yeah. If I'm in town, it's great. I was like coming in person versus. Thank you so much. So much. So it's the other stuff. Appreciate it. I mean, there's so much in energy, but we can talk about this. I'm happy to just let you lead the conversation however you think it should go.

What's going on? How should we even be thinking about oil prices right now, the impact of the war in Iran? How do we, how should we be processing this news? Okay, so we did this. So there's a million things to comment on, so I'm going to try to segment what I comment on, but we did get involved militarily in Iran, which I would just say in my non-area of expertise, in general that is a good idea. I think it's a better idea to have Congress involved, if you can get them involved. And I think it's a good idea to have a lot of expertise on international oil markets being brought to bear when you do this. I think this administration does a lot of good stuff on energy, but I think it's pretty clear at this point that there was not maximum expertise brought in on this. And so the standard issue when you're talking... And indicators there would be like the strategic petroleum reserve hadn't been refilled.

I mean, we can talk about that. So let's make sure to get back to that one. But the main thing is just, you know, rough numbers, you know, straight of Hormuz, which Iran has control over, is 20% of the world's oil production. is flowing through that every day. And so whether you're looking backward or looking forward, there is no replacement whatsoever for opening that straight and keeping it open.

And one of the things we've heard from different people is, oh, we have a lot of options on the table. Like there's a lot of things that we can do. There are definite things that you can do, but none of them is 20 million, right? So it's just, I mean, that's more than U.S. oil production, just flowing through that one place. So, yeah, we can talk about the other things,

but I think the first thing people need to recognize is if you do not get that thing open, then you have dramatically higher oil prices. And then the other thing to get is oil prices really matter. Straight to Hermuse. Dumb question, can't you just go around? I mean, you can try to some extent. But there's a reason why 20 million a day are being routed through. Because it actually comes out of there as well, right?

Yeah, there's just, I mean, all of these things are very, very optimized. One of the thing about just understanding the energy industry is like things are just very optimized in terms of where the infrastructure is. You're doing all of these things. So there are other, and we'll talk about this. I mean, there are certain pipelines you can pipe more oil through. There are alternate routes, but we're talking about the millions a day, maybe. Most of our other options are sort of in the one to two million barrels a day.

And just so people know barrels, 42 gallons, so you're talking about almost 100 million gallons a day for the stuff. But, yeah, so you don't have a lot of great options. And I think that's very important for, I'm going to give some options. but the number one option is keep that thing open, and there are basically two ways to do that. One is you just win the war, if you want to call it that a lot more quickly.

Like if you get some form of surrender and you have friendly people controlling it, then guess what? It can open. That's one thing. So I'm not an expert on how to do that, but schematically, doing that is very effective. The other thing, which is a little bit... But game theory for Iran is they actually benefit,

from global markets being in turmoil because it gives them some leverage over any type of negotiation. This is the biggest damage that they can do. I mean, clearly we've seen they're not going to do, you're not going to send missiles to Israel and get rid of Israel, let alone do anything to the U.S., but they have this incredible control of one of the centers of the world economy. And they have a bunch of stuff going on there. So let's talk about that.

So option one is you get a surrender and they don't use all of their options. But then let's talk about their options. So they have mines in this thing. So they can blow up mines. They have missiles. Maybe the worst thing they have. And this is a change in recent years. And you've had Palmer on here and Ethan on here.

The drone thing is just a total game changer because you can have a small vehicle. You can launch one of these shaw heads out of the back of a truck. They can go really far. And that's a lot harder to deal with than a fixed installation. Yeah, they have something like 500 miles of range on a typical... Yeah, depending on... Yeah, right. The truck drives up to a beach, drone launches,

as soon as it finds that cargo tanker that has a bunch of oil on it, it hits it. Yeah, and then there's the... You're worried about that as you're going to go. As you're going through. So you have to think about those are the kinds of threats you face. So how do you deal with that? I mean, basically you need the U.S. to lead some sort of convoy where people have sufficient security and economic assurances that will go through.

So what's involved in that? Well, one thing is, if possible, you want to get allies involved. And it's even possible you get unconventional allies like China, who has, well, they depend on this. Okay. Now, they have a bunch of, unlike us, you want to go back to SPR, they've been filling up their reserves. So they have more reserves relative to their imports than we do relative to ours. If you think about ours, we have about 400 in the SPR, and we can release about $4 million a day. So we don't have as much.

But still, they really care about this and a protracted thing is bad. But for sure, Japan, South Korea, India. So one thing is just at a high level if you're doing this convoy, can we get these countries that are aligned economically, maybe China, but definitely these others, can we get them involved? Because then they have military presence there. That's an additional threat to Iran if they attack one of their ships. But that's a macro thing, is just can you do this alone

or can you do it with allies? I think ideally you would do it with allies. Then there's a question of what you do about these various things. And I think the biggest, I mean, the other macro thing you can do is just in some way credibly threaten Iran and say, hey, if you, and I can't give the details of how to do it, but if you attack us, if you attack in the Strait of Hormuz, it's going to be very, very bad for you.

So just at a high level, can you make that threat? So they understand this is going to be very, very, no, you know, we've wiped out a couple layers of leaders. ship, so they might be taking these things seriously. After that, I think the biggest thing from what I've heard for military experts is just these drones. Like, what do you do about these drones? How do you give any kinds of assurances?

And as far as I can tell, we don't have any one perfect solution, but you can do a certain amount of stuff from the air. You can do a certain amount of stuff from the ground, although it's very, very expensive. And then most cost-effective where you can do it is you can take out certain stockpiles and facilities with the provisive of these. things are decentralized. And of course I have no specific knowledge about where they are. But in theory, if there's the suicide drone, the Shahid, there's a factory and you take out that factor, you have reduced that capacity. So that's going to be it. And you know, I know you guys

had on Ethan Thornton, I'm friends with. And, you know, he talks about this a lot is just this cost asymmetry issue, is you just want to be doing things. Even if you're way wealthier, you do not want to be spending $3 million to take out $30,000. Which might be what happening with a Patriot missile battery. This, this, this, yeah. that's the kind of thing that happens. So you need a combination of those kinds of things. And then the other thing, so that's all the security stuff. So you get the allies involved.

You take these actions specifically against drones. I think the, what was I going to say about the other thing you can do? I'm just making sure. They're still just. Well, like the other big factors. Oh, the insurance. Okay, insurance. Yeah, the insurance.

So, and this has been floated by the president. And it's a reasonable idea. But my understanding is the insurance vehicle we have, the development corporation, does not have enough funding, so you might need to go to Congress. But basically you want to be able to say, hey, we're going to lead this as the United States. We're going to lead it with allies. If Iran does anything, it's big trouble for them. And we can counter their specific attacks.

And if something goes wrong, you're insured. You're trying to create that confidence to get enough people going through, and then you get a certain amount of stability, and then you start to get closer to your 20 million barrels a day. Either that or you just defeat them really quickly or both. So the other factor is just the loss of production, disruption to production, plants shutting down, plants being damages, damaged. How does that factor in? Do those, do those, do refineries get brought back?

And I'm talking about an allied, allied country. You mean just the trend over time or what? Yeah, generally, I mean, just seeing the videos coming out, you've seen places, I think, like Qatar, Kuwait, places. places, like, yeah, basically just saying, like, yeah, we're going to pause production because we don't even have a place to store this. And or they're suffering damages. And it doesn't make sense to keep, you know, refineries online. This is all, but this is all just downstream of do you have the root open? Yeah, yeah. Like, if you have the root open, then you're good.

It solves everything. If you don't, the way to think of the other things, so we do have some other interesting options. We have to think of them as these are temporary stop gaps. Like if anyone says, I mean, you hear some crazy things. People were like, oh, Venezuela. We talked about this last time. Venezuela is less than a million. I mean, it's beyond crazy.

It's just impossible. It's not like Venezuela has the ability to just increase their oil production by any significant amount in the near future. I mean, we're talking about companies considering going in harm's way to get some incremental boost. So Venezuela is essentially useless right now. So best case scenario in the next year, Venezuela goes from under a million to two million? No, I mean, two million would be insane. I mean, if we made it our entire goal in life, I don't know. But why would you for that?

Venezuela was it like $3 million? Yeah, yeah, yeah, some years ago. Yeah, yeah. Right, but there's so many. But there's a lot there. It's not going to happen. It's just so, yeah. So Venezuela, I would not even, that's just in the category of, it's not even a bad idea.

It's just not an idea. Sure. There's nothing. There's nothing there. No, no. So the bad ideas, by the way, are the worst idea imaginable is ban oil exports from the United States, which I have heard floated in conversation. Seems somewhat reasonable.

I want the oil. Yeah, of course, because we have oil, so we don't want to sell. So, I mean, so many different things. But one obvious one, not obvious one, but it's obvious if you know how these things work, is the refiners in the U.S. do not match up well to the oil produced in the United States. The refineries in the United States are based on heavier crude. Like crude is rated in terms of weight and sourness. Okay.

And so our refineries are mostly for heavier crude. It's actually one reason why Venezuela has some appeal. Canada has a lot of appeal. Because they have heavier cruise. Yeah, and we're going to talk about Canada. We're good at refining it. We're good at refining it. But we produce a lighter cruise here.

Yes, because the Shell Revolution was just this dramatic shift very quickly in oil production. So our refineries are not primarily equipped. That's why when we ended the crude export oil ban, I forget it was 2014 or something like that. It was just this huge unlock because it allows the global market. It allows us to produce for a global market. You can't look at American oil production as a monolith. Yeah, it's not. We import heavy crude and export light crude.

Is that roughly correct? That's roughly correct. Yeah. Yeah. So that's one thing is it's just insane to do that. But in general, think about it. What is the value of higher prices? Is it stimulates production?

We're going to talk about all these different ways where we want to unlock oil. So you want to tell American oil producers, hey, we're going to actually totally screw you over. It's actually going to be worse for you maybe than it was before because you don't even have a market. So you're going to strand all this oil. So that's terrible. There's also this financial manipulation idea, which has. has been floated, which I think is terrible.

Hey, let's manipulate the financial markets basically in a way that we're, you're selling future short on a certain time period and then you're buying them. What you're trying to do is lower than near-term price. Oh, like Fed, like Fed policy basically. Yeah. A very American way to try to sell a world problem. People that like that. The only solutions involve unlocking oil in one way or another.

You do not want to screw up the markets so that people are less inclined to unlock oil, nor do you want to screw up the markets to destroy information. So again, the main unlock is reopen that straight. There's no way around that. But then we can talk about other things. So one category is what you call spare or emergency capacity. Okay. And this is there's some uncertainty here, but it...

Quickly, this is separate from the strategic reserve. Yeah, yeah. So, well, emergency capacity is the strategic reserve. Oh, it is. So spare means you could be, you could pretty easily be producing more per day, but you're not because you're not happy with the price. And this is mainly Saudi Arabia is considered exhibit A in terms of amount of this. Now, there's a debate among experts about how much they have.

But some people estimate they have one, two, three million barrels a day of spare capacity. They could ramp up on some kind of. And do they know how much capacity they have and they're just not sharing it? Or does no one know? I mean, that's always been the story. That's been the story of the Gulf, right? They're not going to, if they just flooded the market with every single barrel that they could possibly produce, they would be getting a worse, a much worse price.

Yeah, and there's this cartel arrangement that does. I mean, in a sense, everyone has this on different timescales. You think about U.S. shale producers. If prices were sustained at $100, guess what? There's a whole bunch of shale deposits that they could produce at $75 a barrel and make a fortune on it that they're not going to do four weeks ago when it was around $60 a barrel. But in terms of the thing about the Saudi oil is it's produced at a much lower cost. So this is them saying, hey,

We have some of this on the table, and then the question is you need to be able to produce it, and then you need to be able to transport it. So you have estimates around maybe you could get one to two million barrels a day. So that's one of them. In terms of emergency, if you look around the world, you don't have a lot of immediate spare capacity besides that. So we have what's called emergency capacity or strategic petroleum reserve. You know, it's already mentioned that we didn't fill up our strategic petroleum reserve. I think this is unequivocally a mistake. I mean, if you look at, the whole thing of the strategic petroleum reserve is you want it for when you need it.

And the best possible time to fill it up is when prices are low. And is that physically a place with a bunch of barrels that we store? There's a bunch of different parts. It's like the Fort Knox of oil? Well, there's multiple different places. But you can think of it roughly that. We basically cash the physical oil in America or somewhere we own it or can access it. And it's sitting there going unused until we're ready.

Yeah. Now there's issues of Biden. Biden definitely misused it. So they used it to basically have lower gasoline prices during midterm elections. Oh, okay. Even though parenthetically their entire policy goal was get rid of fossil fuels, which means you want the price to go up so people can't afford it. There's a very cynical kind of move.

They did it in a way that degraded the facilities. So let's say we're at about 4 million barrels a day. We can get out of that thing. We have 400 million in there. We should have over 700 million. We could have easily filled it up. I mean, it was a perfect time when you're talking about $50 barrel. oil, $60 a barrel of oil.

Yeah. Now it's take... And can I do the basic arithmetic of 4 million a day, 400 million in the reserve, 100 days of oil? But that's not actually, if we're talking about like relieving price pressure, you could trickle it out over

a year and have one 20 or one quarter of the effect. But just think about $100 million a day, 100 million barrels a day is roughly the global market. Okay, okay. So 4 million is just, that's its maximum throughput capacity. Got it. It's kind of an analogy to back.

So there's a certain amount, like with a battery, if you hear, there's a one gigawatt battery installation. That means that usually means there's one gigawatt for four hours. But if you wanted to do half a gigawatt, then you could do eight hours, right? It's the same deal with this. But keep in mind, this is not a hundred days of U.S. oil demand. And by the way, for the refinery reasons, we can't just supply it all with R.S.B.R. It's just the maximum output is, but four million barrels a day or something.

more tight, that's one-fifth of what's flowing through the Strait of Hormuz. So, yeah, it's a thing. But yeah, you can think of, yeah, maybe we'd be willing to do one to two million barrels a day from ours. Now, interestingly, the International Energy Agency, we are part of the International Energy Agency Reserve Program. So we have about 400, but they have about 1.400 million. They have 1.4 billion. So we have allies around the world that can also release this oil. So maybe we could get one to two million from them.

Now, at least last I checked the news, they hadn't agreed to do this because they don't think it's an emergency. Sure. But that's what it's there for. Yeah. So notice the pattern is. At what point would they think it's an emergency? I don't know. I mean, and we can put pressure on them.

I think some of these determinations have to be made in conjunction with the military determinations. You can't make them in isolation. Because if you think about what's the military objective, how close are we to that? Or how close are we at least to opening this up via high security? and financially backed convoys, then you can think about the timetable here. If you think about, well, the straight of cornwall is going to be closed for five years,

let's just get used to being poor. Black bill. No, I mean, this is, people don't get oil. Like, oil is, there's a reason why I focus a lot of my life on oil. There is nothing more, there's no material in the world of energy that is more valuable than oil, because oil is just so unique in terms of it, has this very high energy density and portability.

And there's nothing like, I mean, nuclear has higher energy. What about plutonium? Yeah, yeah. It doesn't have the same portability. We're going to be rich. We're going to have energy too cheap to meter in five years. I talk to a lot of nuclear founders. Well, look, I love nuclear as much as anyway.

They don't even have a portability solution, though, for the near future, unless you're talking about aircraft carrier or an icebreaker or something like that. So I'm just saying the world runs on mobility. And oil is ideal for mobility. Oh, portability in the sense of like using nuclear to actually power. trade and commerce and moving ships around. Yeah, yeah, I mean, even if you can power the grid, you will be poor because you will not be able to trade. Yeah, trade is really important.

Trade is really important. Trade is, trade, and well, and also it's within the country, too. We don't have the nuclear trucks to move things around. Sure, sure, sure. Yeah, so obviously you want super cheap electricity and you want super cheap transport fuel. And for, and we want to see if we can electrify some of this stuff. Yeah. If you can do it cost effectively, it's just.

Some people working on like electric bulldozers, for example. It's very hard. And by the way, yes, what, you need a lot of oil to even get that whole supply chain started because you need the mine. Yep. So, bottom line, yes. I meant that thing about we can all be poor very seriously. Very seriously. So we have, if you look at the spare capacity, I think that's all of them.

So we have the rest of the IEA. We have ours. And then to my knowledge, we have Saudi slash UAE, them using spare capacity and then piping it where they can. So maybe combined, we're talking about $6 million a day if you use all of those. I mean, somebody could imagine $10 million. And all of these have time, particularly the emergency reserve things. These all, like at least the Saudi one is you're sort of, you have a deposit, a big deposit. Here, it's just you have a small reserve.

So you can't do this forever. And of course, there's a risk of if you deplete the SPR more, what if you really, really need it in the future? And then there's a bunch of nearer term stuff that you can do. and some of which I like, just because it could be done anyway. So number one to piss a whole bunch of people off is let's stop using this Jones Act. Do you know what this Jones Act thing is? I've heard of it, but explain. So the Jones Act is a set of restrictions that say that you can only transport things among ports in the United States

if you have everything is basically American. So it's owned by an American, it's run. I don't know, all the, it is all this ridiculous. Yeah, it's basically American ship, American crew. Yes. Yes. So it's, it leads to all these. crazy, crazy things. Like we end up importing things from, like instead of importing it from one

part of the country to another, you end up importing it from some really far away place. It leads to that kind of thing. But in this case, in particular, this, as Californians, we should all recognize this. We don't have oil being piped into California. So we want very efficient maritime transportation. And if you suspend the Jones Act, then you can get more efficient transportation, which means we can at least have lower prices here and we can have less of a price shock here. But in general, I want people to get used to having No Jones Act because I want there to be no Jones Act.

So that's a good thing to do. My favorite idea here is... And I imagine there's a bunch of powerful lobbies that will fight and kill for the Jones Act because there's an entire industry built around... Yeah, it's the idea that we can only have a good shipbuilding industry. I don't know what people think happening with our shipping like our ship building industry is not going as well as other people's.

It didn't exactly work. Yeah, I mean it worked to protect certain people's. I'm just saying if you look at American shipbuilding, it's certainly allowed us to be competitive on a global state. The shipbuilders would say we're early, but I take your point. Early, but you've seen, but I'm just saying like how many decades of declines. Yeah, okay. Well, I think John's like, we're early. It's like a smiling curve.

like churn, it goes down and then goes back up. So, that's, that's the, uh, all right, well, it's, it's definitely not going to help us near term. So the, here's the most interesting thing, which I, is a pet issue of mine, because we're, like this administration, I should say, does a lot of good stuff on energy. Okay. I say that I, you know, I'm in a fortunate position where I get a lot of people ask for advice and stuff like that.

And I feel like one of my jobs is to tell people things they're not doing. A lot of things they'll just do right on their own. But Canada is not one of those things right now. We are just absolutely sleeping on Canada as an opportunity. So Canada has no people and infinite resources and is really friendly. Yeah. So they have these oil sands, you know what these things are? Yeah, you know what explain.

So they're just like these deposits of oil. It's like nature basically committed an oil spill is the way to think of it. Okay. So all these oil sands, for years they were not very useful. They weren't viable because it's not just you drill down black gold comes up. It's like buried in the sands. There are different kinds of things, so you can mine it out directly, or you can do what's called in situ, which is you heat it up underground. Oh, and it liquifies.

In any case, they have just this unbelievable oil seep. They have way more oil deposits than we do, way more oil reserves, but they have on the order of a third of the production. And in part, because they have absolutely terrible policies, which are their fault. But in part, we have had terrible policies, including getting rid of the Keystone XL pipeline and this kind of thing. So, but if you just think of Canada, it's just we should be so, they have water, I mean, again, no people, one-tenth of the population, infinite raw materials. They got everything. They got timber. They got natural gas. They got oil. Like, and we're talking about Venezuela. They're a friendly country, right? They got smart people there. They haven't been driven out of the country because they're afraid of getting killed by Chavez or Maduro or whatever. So this is an opportunity to say, hey, Canada, let's work, let's have a task force. You got, are at a low in terms of, we don't have pipelines, but we can do rail. At least we can do rail, maybe trucks, but at least rail. Rail transports from Canada are really low right now. So let's have an initiative where maybe you bring a couple hundred thousand, depending on your

expectations, maybe we can add a couple hundred thousand barrels a day. So that's one that excites me because I want to be, like the US Canada superpower thing is this huge opportunity that we're sleeping on because of the idea of oh, we should only be doing it in quote America. But this is just, by the way, unbelievable, I think they have best uranium in the world. Yeah. It's a maple syrup's up there. Yeah, I don't have a Canadian personality, but I love that place. No, strategically.

It makes a ton of sense. And they're very friendly. Yes. They're very friendly. Maybe we should be friendlier. Yes. You know, no, we should, and we have trade agreements coming up. Yes, yes.

The other thing is we have some, you know, we can increase capacity here, as I mentioned, depending on prices going up. Yeah. So there's that, but that might be a couple hundred thousand dollars a barrel of So high level, open the straight of Hormuz. That's the only thing that really works. Then you can talk about Saudi, Mideast spare capacity, our emergency capacity, and IEA emergency capacity.

And then below that, you have Canada Jones Act and U.S. production. That's a really, really thorough list. Thank you. I appreciate that. Anything else, Jordi? What are you going to be paying attention to you most closely? Do you watch the price chart? Is that important to you?

Well, it's important. And I mean, I'm, it's weird, I work with, I run like a kind of a network of energy executives, but I'm focused on the policy side of things. Sure. So I'm often laughably out of touch. I mean, I know in general within $10, usually what it is. Yeah. But as a policy, I don't, I don't invest in energy.

You're looking at longer term, five years, but also like I'm deliberately avoiding exposure to the commercial part of it. Of course, of course. Yeah, you want to be independent. Yeah, I want to be independent. So, yeah, I'm not. But. at the moment, yes, I am looking at the prices. But I don't do predictions, but I'm just saying, I think there's a service right now and just laying out the options for people.

Because even in the government, I mean, you know, not everyone has a lot of expertise in oil. And this is the kind of thing where, yeah, it's kind of why I think you texted me this morning. I mean, sometimes you need to know about oil today and sometimes you don't. But when you need to know, you need to know, you really need to know. Yeah. Well, it seems like whether you're in the business community or just, you know, humanitarian, everyone should be rooting for a swift and peaceful resolution. Yeah. Do you have any insight as this might be totally out of your wheelhouse, but insight into how people running businesses that are most impacted by fluctuations in oil prices? Like, are they, have they been actively hedging over the last couple weeks? Are they doing, like, how do you run your business? when your input costs can go up dramatically,

potentially at a much higher rate, than you can adjust your own prices? I mean, just in general, these guys are sophisticated, I think, within the bounds of doing it. I don't think you can hedge indefinitely for this big a price kind of differential, like somebody is going to say. I mean, but it's also, you think about

there's on the consumer side, but on the production side is even more dramatic if you just look at the margins of these guys. I mean, the oil industry is just fascinating. I mean, I just remember anecdotally when I came out with my first major book The Moral Case for Fossil Fuels in 2014. Even I could see speaking engagements

correlated to these prices because you had a big price crash around them. And then you think, wow. We don't need Alex. Yeah, it might be nice to keep the troops motivated, but you'd probably rather be employed. That's wild, yeah. But it's a hard, it's a hard business.

And one of these things is you just, the people who survive have a very strong constitution. And they can manage the risk very, very well. I was talking with a bunch of these guys, these guys, and actually people we know in common. And like some of the better ones, some of the two of the young superstars in the industry, they're just saying like, we love the chaos. Because we can just, no, we can handle it. Like when the prices go down. Sure.

Because that's, you think about what they've been dealing with over. the last year, they've increased production, prices have gone down. And you look at this, this administration is very focused, I would say too much focused on having prices go low. This is one of the variables here is, although the administration I don't think was fully, was not fully prepared for this situation. In general, they are maniacal about prices going low. Like Trump in his mind has a very strong focus on $50 a barrel, which I think he feels like that's always the perfect price for oil. Sure. And the people in the oil industry say, wait a second.

had inflation of everything else. Why does our one product have to stay at $50? And that's part of the appeal of Venezuela, rightly or wrongly, because it's not going to make a big difference soon. But it's like, how do we keep oil prices low? But that's hard to be in that industry as a consumer, as a producer, rather, and it's hard as a consumer. So they have financial instruments. But the main thing is you just, you need to, you need to just have the flexibility in your business, in your business model. And it's, it's, oil is again, it's the most valuable material in the world of energy, which is why everyone uses it. But it's also why it's so inelastic and you have these price fluctuations. So you just need to come to terms with that or somebody needs to figure out something better.

But it's really hard. And of course, policy-wise, we can have better policy to make it more stable. But it's still that thing where it's so valuable that guess what? You get a little extra demand, a little less supply. Price goes up a lot. It's different than other things which you can substitute for much more rapidly. You're out of a job when we build the Dyson sphere. Oh, man, I got, I'm working on this AI thought leader thing we talked about last time.

So I got, I got plenty to do. We'll cross that bridge when we come to. I got plenty to do with my time. Thank you. As long as we need energy, you'll be doing just fine. All right, guys. Thank you so much for coming on the show. Let me tell you about Shopify.

Shopify is the commerce platform that grows of your business and lets you sell in seconds online, in store, on mobile, on social, on marketplace, and now with AI agents. And let me also tell you about app loving. Profitable advertising made easy with axon.aI, get access to over one billion daily active users and grow your business today. And without further ado, we have our next in-person guest live in the Ultradome.

Michelle Volz, how are you doing? Good to see you. Good to see you. Thank you so much for coming down to the TBPN Ultrodome. Yeah, no. For those who aren't familiar with your venture capital career and your marathon running career,

introduce yourself. Yes. So I am Michelle. I was previously at Andreessen Horowitz. I just launched my own fund packs, which is a $50 million early stage venture fund focused on foundational categories. So I was on the American Dynamism team, Andreessen Horowitz. Similar themes to that, just earlier stage.

Biggest lessons from Andrews and Hornets. Can we hit the gong first? Oh, yes, we got to hit the gong. I think Michelle should hit it yourself. Here you go. Hit the gong. You're here in person. Go for it.

All that's already sort of broke. You won't break it. Smash that. That's kind of tennis form. That was good. That was good. Anyway, let's start with lessons from Andrewson Horowitz. Like, what did you learn?

What are your biggest takeaways? What are, what's the playbook that you think you will be continuing? And then we can go into what you think might be different. Yeah. I mean, I think in the categories that I look at, it is common for founders to need to raise a lot of capital. Okay. And the ability to be able to fundraise for that. capital ahead of maybe what are traditional milestones and software investing is very important.

You need to be able to be a magnet for capital, for talent, and sell a vision, because what you're doing is very hard with fewer obvious proof points maybe at the early days. And I think identifying that in founders early is something that Entries and Horowitz is very good at. Yeah. In American Dynamism companies, hard tech companies, it does feel like there's like high cap-X requirements, high human capital requirements, you're not going to automate everyone on your supply chain on day one. And so there is this, like, how you talk about growth, how you talk about scaling and success and finding product market fit. It's, it almost feels back to like the, okay, we're scaling DAU and we're, let's focus on that metric instead of like profit in software.

what does it look like in hard tech? Is that the, okay, we have a program of record, or we have these SBRs coming in, or we have this long contract that is very solid, but we're not going to draw down on it for a few years, so we have a capital need. But how do you think founders in hard tech should be talking about what progress looks like at top line or whatever top line means? Yeah, I mean, I think that's the hardest part. In software, there's such a structured set of metrics that everybody has been trained around. Like, gross. growth rate, like weekly active users, revenue. It is so different company by company in these sectors, and that's why the storytelling is important

because if you're talking to later stage investors, they might be trying to make a pattern recognition off of a pattern that doesn't quite exist yet. And so in defense companies, like, yes, there are government contracts you should be trying to get, you should be working towards programs of record, but they're like different intermediate proof points before that. I usually explain it to founders as like the three things. you're trying to make progress on are people, product, and then, like, revenue or traction. And so it's like, are you hiring the best people and the people that are experts in the different

categories that you need. So maybe you need a mechanical engineering hire. Maybe you need a government sales hire. Maybe you need software as well. Are you making progress on people? Are you making progress on product that's tangible? Like, can you show that there's, like, developments? Maybe that's, like, you've built something and it's, like, got into a test milestone.

Maybe it's in the hands of users. like are you making product progress? And then are you getting some sort of signs from customers that like they are willing to pay for this? Like are you getting contracts? Are you getting L-O-I's? Like are you getting something that shows that there's a commercial use case for this? I think sometimes in like hard tech, there's a million names, hard tech, deep tech, frontier tech, American dynamism. I think sometimes people skew more towards thinking it's just deep tech, which is like I would describe it.

Exactly, like science risk. Like nuclear fusion is probably like the most. extreme example. That's hard. It's like, of course there's a customer if you can do this, but like, we'll see what the timeline is, hopefully soon. But I think in a lot of the categories I look at, there are commercial proof points you can make.

Like there's usually a latent demand or like, or an immediate demand. It's like, how do you show that you can get those customers and they can like signal that they will buy from you? Yeah. There was this viral Satrini article that said software is dead, sell everything. thing. How are LPs thinking about hard tech as a category with that backdrop of like the SaaSpocalypse? I imagine it makes things easier for you, but has it actually rippled through the LP community to the degree that your investors are more excited than ever to have a hedge

against chaos in automatable software, all of that. Yes. I will say this with love to all of my LPs. I think founders are typically like leading indicators in categories. Like they, they are at the cutting edge, like they are seeing around corners. Often like great founders will see something before VCs do. Like VCs are lagging indicators. Like once something is hypey, it's like usually too late to find a good early stage investment. LPs are lagging, lagging indicators. Sure. That once something is very mainstream and there starts to be some signs of returns, then they're like, we should pay attention to this category. And so I would say, A thing that was more surprising, partly because I live in my own echo chamber and constantly hear about all of these great companies building and, like, defense tech, hard tech, and, like, industrial tech.

LPs weren't as familiar as I expected. So, like, there was a level of education. I think there's still some weariness of, like, there haven't been a ton of exits yet. I mean, I think that might change of SpaceX IPOs. Yep. It would be one of the biggest IPOs ever in history. but I would say like the the appetite for it grew over time as I was fundraising. On SpaceX, yeah, I mean, great point.

It could be like $1.75 trillion is the number that's being floated. It's staggering about it. It's going to return like not just like one venture capital fund. It's going to return like 20 venture capital funds. Like everyone who invested at early stage is going to see their fund return from that. But all the LPs had to go on a 20-year journey with like a million, continuation vehicles. How are you thinking about timeline messaging to LPs? Because 10 to 12 years is what I've heard recently is kind of standard. But I've talked to some VCs and say, yeah,

like most of my LPs want their money back in like six to seven years. And yet in hard tech, it feels like it might be longer. Are you communicating a particular timeline or is that even a relevant conversation? Yeah. I think like every, every fund roughly pitches 10 to 12 years. That's sort of the expectation. I think it's actually, and it's actually, interesting catch-22 because if you had ownership in SpaceX, your LPs actually like hold on to that. Yeah, totally. Like keep that. I think it's going to keep going up. But maybe you would have an option to take some off the table and return some, you know, money to LPs if that's what they wanted. But often if it's like a really, really good deal, LPs are like, oh, I think it's

going to keep going up. But I think it's like case by case. I actually just was talking to LPs about this and they were like sometimes it makes sense and sometimes like hold on and I think if you build a lot of trust with your LPs like they're usually along the ride with you what what GOs are you spending the most time right now obviously the it feels like we have so many so many founders that are just like in in our kind of neck of the woods here but are you spending a bunch of time in Texas Bay Area where where do you expect to deploy most of your dollars yeah I think like there's five GOs and it's San Francisco LA Austin, New York, and D.C.

I'm actually seeing a significant amount of companies in D.C. Austin, I would say, though, is the biggest surprise for me. I know people have been trying to make Austin happen for a while, and I was skeptical, and now I think it's sort of been an explosion of great companies. It was kind of the same thing with L.A. Like, there was this long L.A. movement a while ago. Snapchat was sort of like the first big, like, you know, public tech company or one of them to come out of L.A.

and then the El Segundo thing just happened like completely organically without any of those people that were set cheering for L.A. And it just sort of popped up and it seems like Austin's going through a similar thing. What's unique about the companies that are building in D.C.? Are they thinking more demand generation first, more work on Capitol Hill first? And then maybe they'll set up a satellite manufacturing office or work with like a contract manufacturer to deliver their capability. Yeah, I think like you're seeing founders who are working at companies like Palantir, for example. that have a big office in D.C. They've been relocated to D.C.

They're close to their customers, and then they spin out, start a company. Usually these companies will end up opening an office in New York as well. It's just a deeper pool for engineers. But there's a surprising amount of really strong engineering talent in D.C. That is pretty fixed. They want to stay in D.C. They have a life.

They have a house. They have a family. So it can be a very good place to start a company. What's the view on shipbuilding? Jordy says it's impossible to build a ship in this country. He was saying it can't be done. Is it possible? We need to figure it out.

The press are going to be behind. I think, I mean, like, the optimistic take would be like we can make it possible. We should make it possible. The pessimistic take is if we don't make any changes into the regulatory environment or the cost of building these things, it might not be. And that would be a shame. Yeah, that makes sense. Strategy with the fund.

Yeah. Are these like $2 million lead, pre-seed type checks? What's the deal? Yeah. On average, one to $2 million checks, like early stage, try to be in like the pre-seed seed occasionally a little bit later rounds. Like to partner really closely with founders. And I think the special thing about the early stage,

which is just my favorite stage of company, is you get to be very aligned. It's like a significant check for me if I put, you know, one to $2 million into your company. and I'm very motivated to help you be successful to get to the next round. And then for that next round, I can be much more strategic in helping you get there because I can't lead your next round. Whereas I think sometimes at the multi-stage funds, it can be a little bit tricky to give advice for future fund raises because you might be able to participate.

And so you have to be more delicate about it at a big fund. And as a small fund, you just get to like be totally aligned with the founder. One last question about sort of the category targeting. American Dynamism made a lot of sense as like a theme around defense tech and hard tech, and you put a manufacturing company in there. One of the interesting subcategories of AD that I always popped out to me was something around education, like re-educating workforce education, also just like education technology broadly. Do you think that that's within your purview or is that something that,

you're sort of carving out and focusing more on the deep tech, hard tech, physical products. Yeah, I would say I like to look at things that sit at the infrastructure layer of society, which could include education and housing, sometimes even healthcare, often things at the intersection of hardware and software or tech and government, or just like doing something fundamentally important for like the broad majority of Americans. I think American dynamism is this beautiful phrase. That can be an umbrella phrase as well. I sort of embrace that as well with PACs.

But at the early stage, what you're really betting on is people. There is so little to understand except the team in a dream. Are you building something in a category that's very exciting? Are you taking the big swing? Do you sort of have a right to win in that category

and some sort of unfair advantage? I like founders that have worked in the industries that they are building in before. And do I believe you can be that magnet for capital and talent and everything? Yeah, where are the big talent pools these days? There's the SpaceX Mafia, the Palantir Mafia. It feels like there's an Anderol Mafia now. Probably going to add up a base power mafia in Austin eventually. Yeah, I think base, I think Serronic is starting to see a few things. Basically any company that's, that's like reaching escape velocity or like escape velocity and product market fit has grown really quickly and has hired exceptional people.

There are some people that really love that zero to one stage and the company gets big. I think we're seeing it at Andral. We saw it at Palantir. We continue to see it at Palantir. People learn how to grow quickly and then want to apply that to their own company. Amazing. Well, thank you so much for joining by. Thank you.

I'm sure we'll have a rest of your day. Let me tell you about Vanta, automate compliance and security. Vanta is the leading AI trust management platform. And let's go over to some hard tech that's happening with Stargate. By the way, John, apparently Trump said something to the effect of, I think the Iran war is very complete, pretty much. Oh, okay. So we may be on the verge of new era of peace. Crude is back down to like pre, like Friday, pre-weekend prices.

What do that mean? 80-90? 83. Okay. 83. Okay. Well, maybe he was watching. Maybe he was watching Alex Epstein. And Alex said, hey, we got to wrap this up.

We got to stop. The price is too darn high, says Alex Epstein. Oracle and OpenAI allegedly dropped their Stargate expansion lease, meta swooped in the same day. Ben Pouladeen says, that's not weakness. That's a hot real estate market. And then he goes on to comment that the real story is Open AI passed on six additional buildings because power won't be ready for a year. The power is is is the gating factor here,

not the chips. Interestingly, we are in the power the power bottleneck, the energy bottleneck. And by then, Rubin replaces Blackwell. So why are you going to build a Blackwell data center when you could build a Rubin data center? So why expand with last gen chips? The eight building, 400,000 GPU base build continues. This is smart capital allocation. he says. And so this is based on reporting from the information about Oracle and OpenAI, sort of adjusting the plan for Stargate. I am particularly interested in digging into the Stargate of China. They have a project that's the Chinese Stargate, funded to the tune of $36 billion, not up at half a trillion, but still, it's a very big number. It's a big number and it's growing. And I've been,

I've been digging into it a little bit, trying to learn as much as I can. We'll do a deep dive probably with someone at some analysis who knows a lot more about it. And chat, GBT is back at the top of the App Store leaderboard. You know, we do have to... Adam, GPD says GPD 5.4 cents. It's for gods. You know what that meme is from? Big Chungis.

It's a good one. Anyway. Yeah, so I'm kind of surprised by this for the one reason being I expected Anthropics been spending a ton of money on marketing. They started ramping it up. Yeah, yeah, I see. It was all of my. Super Bowl.

I expected them to basically be willing to spend whatever it took to stay on top. Yeah, but, I mean, it's capital war. So both sides are going to spend a lot of money, right? And, you know, there was, what? What's up? Wasn't the rise in the Claude, like, ranking just because of the DOW news? And now that's, like, not on the front page. So, so, yeah, that's what I was getting.

It was like, when the Super Bowl ad dropped, and we had some harsh words for the nature of the characterization around ads, our favorite thing in Claude, led us to launching Claude with ads. But our question was, will this Super Bowl ad move the needle for the consumer app? Like, a Super Bowl ad is a massive campaign. Millions, tens of millions of people will see this? It went viral. Like, it's, will it work on people? Will people see this ad that's like sort of deep in the weeds around how ads work in LLMs? Like, will this be enough for people to go and hit the download button. There's a hurdle there. Will they do it? And they did it a little bit. I think Claude went from like like 26th to 18th or it started

climbing then and then you're right. The Department of War narrative really, really shot up. But as with any story in tech, it has to go back and forth and back and forth. And the memo that came out on last Wednesday probably sort of reset the narrative there and changed how people were thinking about that. And also, these charts are based on momentum. And so it's very, very hard to stay a lot. Yeah, Claude had gotten to something like a million downloads a day, according to Mike Krieger. Oh, yeah. And at the same time, Chad Chhabiti last month, I think did like 55 million downloads. So it's still not outfacing. Sure, sure. Well, Paramount Plus is number four in the free apps. Although, how can it be free when you have to subscribe? Sort of odd. Yeah, that's the whole thing.

But bullish for the Ellison team. Chris says UMF's ruined agency so bad that I'm going to have to start saying gumption. Gumpion. Gumpion is really good. I like Gumpion. We hire for Gumpion. AI Gumpion. AI agents? We only hire for gumption here.

What's the noun version of gumption? I don't know. A friend of the show, Amjad, said not having coding experiences becoming an advantage. And sure, yes. says not knowing how to cook is becoming an advantage, CEO of McDonald's. I love it. I'm glad to have some great takes, though. This is just an easy, an easy riff.

It's all in good fun. What else is going on? Lots of debate over the singularity pricing, as Will DePue put it, because barely AI, which I believe is Trunk Fan, says cursors, internal analysis shows how hard anthropic is subsidizing Claude Code. year, a $200 monthly subscription could use $2,000 in compute. Now the same $200 a monthly plan can consume $5,000 in compute. It does feel like there's a capital war. I'm very interested to dig into the history of capital wars in winner-take-all markets, but also in oligopoly markets.

You should write the book on capital wars. Capital wars are fascinating. John Cugin history of capital wars. I lived through it during the Uber and Lyft era. There was a flippening for a little bit where Lyft was at the top of the charts. The friendly ride share company. It was very friendly. The cars would pull up with a pink mustache on them and it had a very, very fun

whimsical branding and you would sit in the front seat of the car because you were friends with the driver. Did you not know this? Did you not know this? This is a real thing. Yeah, so when Lyft launched, people see Lyft and Uber as like commodities at the time but Lyft was like revolutionary and people were super black bill on Uber because Uber at the time was only black cars. So they didn't allow anyone to just hop on the network. Like it was not a three-sided marketplace or whatever, two-sided marketplace.

Like you had to have a like a livery T-CV number. I forget what it's called. But like you had to be registered as a limo driver and you had a escalade. And normally someone would call you and say, hey, there's a wedding and we want you to come shuttle people around at this wedding. And they would pay you for that. And Uber went to those folks and they said, hey, you are registered to legally drive this car as a professional driver. We want to book you through our app. And so the black car became Uber's main product, their only product. And then Lyft figured out through some interesting, you know, strategy, how to allow people to just show up with a Prius or a Toyota Camry or Corolla or whatever. And so drivers on launch day, I remember like the launch week in San Francisco would just show up in just some random car.

But Lyft gave them all pink mustaches, huge pink mustaches. The team can pull up a picture of the huge pink mustaches on the front of the Lyft cars that were very whimsical. And it was remarkably cheap because Uber had been somewhat subsidizing the black cars. They hadn't been taking a huge margin. They were losing money. But the Lyft cars were way cheaper because someone was just showing up with like, their Toyota Corolla or whatever with their pink mustache on it. And so you would hop in the front seat because this driver was your friend.

There you go. The pink mustache on top of the, what is that, a CRV or something. And so you would hop in the front seat. They would ask you to pound the driver. Give him the knuckles. Give him knucks. Give him knucks. This is a thing.

No way. I'm almost sure. So you'd be like, what's up? And then you would get in the front seat and they would drive you. Of course, if you had a group, you'd be in the backseat. But then. Everyone had to give nucks. I think the knucks might have been optional, but encouraged, recommend it because it was like,

hey, we're buddies, and we're fighting the war on the bork. We're going up against. But you were encouraged to get in the front seat. Front seat, front seat in the lift. In the lift, when you're taking a lift, your friendly lift driver shows up, he sit in the front seat. You give him the nux. This is the thing. Pink mustache. Chungis. Big Chunga Sends his regards. All right. You got to hire a car, but you have to be absolute boys. You do. But, but truth be told, like a lot of the early lift drivers were, we're, like super cool tech forward San Francisco people and so like you'd have

a lot to talk about because like people were like they it wasn't professionalized at all yet it was very much like like two-sided marketplace you'd make a lot of money as a driver like there was no margin compression yet it was like free money everywhere for everyone and it was a capital war and Lyft

raised a ton of money and gave and came into the market with a disruptively cheap product and wound up delivering a really fun experience that was very very cheap and then Uber sort of fast followed and launched Uber, what's it called, the normal one, not black, Uber X. Uber X. Uber X, yes, Uber X. And Uber X was bring your own car. And then the rest of the game played out. And it became a capital fight. And for a while, like the meta was, oh, you're, you're, you need a, you need a ride.

You open Lyft and you check the price and then you open Uber and you check the price. And people would check both prices constantly. And then over time, like the liquidity and power loss stuff came into it. John, do you remember Uber helicopter? Yes, that was the partnership with Blade. I think it was a partnership with Blade, which was the Uber for helicopters. And it was a thing, but in like, you know, for a few weeks or something. Did you ever do it? No.

No. Because Uber's had a whole bunch of different, like, stunty products, like Uber, you know, Uber Yachts or Uber. They did Uber ice cream where you could get ice cream delivered. That was like the precursor. to Uber Eats. They did an ice cream delivery truck where they all they they, they every Uber driver checked in with the, with the field office, got a cooler in the back with a bunch of dry ice and a bunch of ice cream and you could press a button in for like two dollars. They would bring you an ice cream. They even did I think Uber kittens and you could ask. Yeah, yeah, Uber dogs or

something. You could like, yeah, yeah, I remember that. Yeah, it was very funny. Anyway, lots of stunts, lots of ways to get attention, go viral, and that was sort of the launch strategy. They would do that for whenever they'd launch new markets. They would come in and they'd be like, we need to go viral on the local news. We need to let the town know that we've arrived. And so lots of back and forth in the capital wars of 2023, 2024, it was fun times. I lived. I fought through them. What else is going on? I think that's a good place to call it. Well, thank you so much for listening. We will see you tomorrow at 11 a.m.

Sharp. Leave us five stars on Apple Podcasts and Spotify. Sign up for our newsletter at TBPN.com and have a great rest of your day. It has been an honor and a privilege. And I hope by the time the show goes live tomorrow, the war is over. Me too. Me too. That would be I would love that. That would be quite nice. Thank you. Thank you. Goodbye. Cheers.