Morning Brew Daily - Tech Stocks Rocked by Global Selloff & Meta CTO Andrew Bosworth Talks New In-House Glasses

Episode Date: June 24, 2026

#874: The US stock market suffered a broad sell off led by tech and AI stocks. In NYC, a political race is underway where sweeping AI regulations are at stake.  Netflix launches a horror game that can be played on your TV. Finally, a special conversation with Meta’s Chief Technology Officer Andrew Bosworth to discuss the launch of the Meta’s new in-house smart glasses and why they think the specs will be an everyday AI device rather than a niche tech product.  To learn more visit https://www.servicenow.com Subscribe to Morning Brew Daily for more of the news you need to start your day. Share the show with a friend, and leave us a review on your favorite podcast app. Listen to Morning Brew Daily Here:⁠ ⁠⁠https://www.swap.fm/l/mbd-note⁠⁠⁠  Watch Morning Brew Daily Here:⁠ ⁠⁠https://www.youtube.com/@MorningBrewDailyShow⁠ Learn more about your ad choices. Visit megaphone.fm/adchoices

Marketers, tell us if this sounds familiar. You invest in something that seems incredible like millions of views, but then don't see any revenue. Instead, invest in what looks good to your CFO. LinkedIn Ads generates the highest row ads of all major ad networks. Spend $250 on your first campaign on LinkedIn ads and get a $250 credit. Just go to LinkedIn.com slash MBD. That's LinkedIn.com slash MBD.

Terms and conditions apply. Good morning, Brutjali Show. I'm Toby Howell. Today, New York's primaries last night showed AI money is going to be a force in the upcoming midterms. Then the market sort of threw up on itself yesterday as AI stocks hit the fritz. It's Wednesday, June 24th. Let's ride. Good morning. You might be wondering, hey, why is this episode so long today? And it's not because Toby went on an extended rant about a new running shoe. We're trying something new, tacking on an interview to the end of the episode. episode. Yesterday, as you'll hear in a bit, Meta released its latest smart glasses, and we had the opportunity to chat with the person spearheading the project, Meta CTO Andrew Bosworth. Bosworth, has been Mark Zuckerberg's right-hand man for 20 years, so he had some very interesting

things to say about this glasses push, as well as the broader cultural problems facing the company right now. We talked about the glasses and what it was like working alongside Kylie Jenner to design a pair. Apparently, she was very hands-on, but we also spoke about morale at Meta, which has not been high recently due to layoffs and AI fears and reorgs. We hope you enjoy this venty-sized episode of MBD today. And now a word from our sponsor, ServiceNow. Toby, are you sick of doing homework for an AI that was supposed to handle the parts of work you hated but just repeated problems back to you?

Yeah. How'd you know? Because you've been complaining about it for weeks. Oh, my bad. Thankfully, ServiceNow's AI specialists are different. They aren't a tool. Think of them as digital teammates who can actually. do the work. They can resolve cases, process requests, close loops without generating extra work for you.

When you can truly delegate to AI, you can get back to the work. Only you can do the work that requires a person with ideas and judgment and, you know, a pulse. To put AI to work for people, visit servicenow.com. That's service now.com. Well, the market is making the sound a balloon does when you let air leak out of it, a sort of as a tech sell off-grip global markets yesterday, causing some of the biggest AI names to slide. It started before even the MBD crew gets up in the morning as traders in America woke up to the South Korean Kaspi Index going haywire. The Kospi plunged more than 10% in minutes,

leading to a trading halt as two of its biggest companies and key AI memory suppliers, Samsung and SK Hynix, both fell more than 12%. Then, like Squid Games and K-pop Demon Hunters, South Korean influence spilled over into America as the NASDAQ opened the day down more than 2%. While stocks in the U.S. cloudback, some of their losses throughout the day, AI memory and chipmakers like Micron, AMD, Intel, and Marvel fell between 6 and 13%.

It's clear investors are taking a minute to reevaluate how much AI optimism is already priced in. There continues to be jitters around how much tech giants like Google and SpaceX are spending on AI, plus the potential of a Fed rate hike looms in the near future. Neil, by the end of yesterday, the tech-focused NASDAQ 100 finished down about 3.3%, which certainly dinged portfolios and led to questions on if the AI-induced rally has stalled out. Have you ever heard what a balloon letting out air sounds like? I could have finished with like a but I didn't do that. And we're all grateful you did. No, the way I look at this is the stock market is the Baltimore Ravens and AI companies are Lamar Jackson, their star quarterback.

When Jackson is healthy, the team thrives and he's playing well. when he's injured, they look awful. And that's because this market is so concentrated in AI names. Around half of the S&P 500, overall market value, $67 trillion, is captured in just 41 stocks tied to the AI boom. That's according to Bianco Research. And the 10 largest stocks in the Vanguard U.S. Total Market Index ETF. So this is pretty much every stock on the market, 3,500 names represent 35% of its value. Top 10 largest stocks of 3,500 represent 35% percent. percent of its value. So the way AI goes is the way the market goes, and we're just seeing a lot of volatility recently. This was a six straight day that the NASDAQ moved more than 1% in either

direction. So it's a roller coaster that seems to always start in South Korea because they are so dominate in AI from those two companies, and then it filters across the Pacific to the United States. Yeah, let's head back to South Korea, actually, because it was a wild day. It wasn't just profit taking. We were seeing. It's clearly become a very recent. retail-driven market over there. And not just that, but it is a leverage-driven retail market, which is a dangerous concoction. Margin debt hit a record $25 billion this month, mainly concentrated in normal everyday traders over there. So when that happens, plus you throw in this thing called a leverage ETFs, leveraged ETFs that are tied to Samsung and SK-Hinix, amplify every move.

So instead of just getting a 10% move in either direction, you get a 20% move in either direction. and that caused a massive wipeout. And that's why we saw, you know, this exchange halted again. Trading volume has been 52% above normal in the past few weeks over in South Korea. And South Korea's circuit breakers have now been triggered four times this year versus zero times in all of 2025. It shows just how frothy it is. So for people don't necessarily speak Wall Street or Cosby, this means that people are, everyday people, retail investors are taking out debt to, fund stock investments. So that's what's happening over there. It happens over here a little bit,

but over there, it's like at another level. One of the themes that I'm looking at right now in the stock market is the trade that's quote, sell the spenders. So we had these huge hyperscalers, like Oracle, Microsoft, Amazon meta, spending collectively over half a trillion dollars on building out the AI infrastructure. And the question is, will they ever see a return on their investment? And so far in the past 12 months, investors have gone a little spooked by all. all this spending. Microsoft and Meta are down 30% over the past year. Oracle, which is basically the biggest hyperscaler outside of the Mag 7 stocks, is down nearly 50%. These companies, like retail traders and Korea, are taking on a lot of debt to fuel these investments. And investors are really

skeptical at this point. The ironic thing yesterday was a lot of people started drawing comparisons to the dot-com bubble. That has been something that has been happening a lot recently. But the timing was notable because it was right after the death of former Fed share, Alan Greenspan, which we spoke about on the show yesterday. They were invoking his famous irrational exuberance line that he wrote right in the late 1990s tech bubble. So you can absolutely argue that we are in a different moment right now because this AI boom differs from that moment in time because these companies' profits are actually growing. The underlying businesses are doing well. But you could also say there's a catch to that as well because even though tech

earnings have been very good. They've been blockbuster, honestly. That spending increases has not been reflected on their books yet. And maybe in future quarters, it will look a little less rosy as all that cast is plowed into data centers. Yeah, I mean, with the theme of the market has been periodic fears of AI bubble. The other theme has been by the dip because it's always worked out for the past couple of years. And if you look at the stock market this morning, things seem to be green. The South Korea index has bounced back. Everyone's looking. at Micron earnings, which come after the bell. Remember, Micron has been at the center of the memory chip boom. There's a huge memory shortage in micron. A stock has gone haywire over the past

couple of months. So this could be some profit taking ahead of micron earnings, but this should dictate the stock market for the rest of the week. Moving on, AI is crashing election day. Yesterday, Democrat Michael Lashar prevailed in a New York City congressional primary that received over $20 million from AI-linked Super PACs. It's a sign of things to come at the midterms this November, where the question of AI and data center regulation will be a key issue, and AI interests will open up their wallets to help elect their preferred candidates. Much of the AI money in the New York race was focused on beating and supporting one of Lashers's opponents, Alex Boris. Borris is a former Palantir employee who's become a

vocal advocate for reigning in AI. As a state assemblyman, he led a bill through New York's legislature last year that requires major AI developers to publish safety plans to the the public, one of the strictest statewide AI regulations on the books. And he said he would try the same at the national level if elected. AI boosters sprang into action. Think big, a Super PAC affiliated with Open AI investors like Andrew Sin Horowitz, spent at least $8 million to oppose Boras. But another Super PAC, public first action, spent $11 million to help him win. And here's where things get very juicy. Public First Action, which supported Boris, has received $20 million from Anthropic, the AI giant. So it's not as clear cut as the AI industry versus AI critics.

This race was essentially a proxy battle for the Open AI versus Anthropic rivalry. Open AI vouching for candidates who want less regulation, fearing it could harm innovation, and Anthropic opting for those who want more rules. It's why some called this primary the AI Civil War. And while it started in New York, the battlefield goes national this fall. Yeah, Democratic rep who is involved in the race. Pat Ryan called the race an opening salvo. His argument was that AI's become the defining political issue ahead of the 2028 election, similar to, you know, on the same

level as tech regulation, climate, or healthcare. This is now the new ideological dividing line that is separating candidates from each other. But this is such a nuanced thing because it is not just pro versus anti. As you mentioned, even it's not anthropic versus open AI because both of them want similar things, but in different directions because their policy position if you're Open AI is that they support a national federal AI regulatory fame work. They just want to know the rules at a national level. Anthropics position has been more of a state-level AI regulation. And those two slight differences in position led to millions of dollars flowing to different

candidates who supported, again, almost the same bill, but just from a little bit of a different perspective. So you're absolutely right to call this a proxy battle for a much wider battle. but it's not as clear ideological divisions as you might expect. Yeah, Michael Asher in his victory speech, he tried to sidestep the whole thing because he apparently didn't ask for all this AI spending. He said, I have some news for the two big AI companies who've taken such an unusual interest in who won this congressional seat.

And I think he's referring to Open AI and Anthropic. I won't be taking my cues from either of you when it comes to protecting our kids, our jobs, and our environment. He was actually a co-sponsor of the Rays Act, which is the AI Safety Bill. that was pushed through by Boris. Now, I do have a inkling of why they took such an unusual look at this particular race. This is the 12th district of New York, and it encompasses Manhattan, like most of Manhattan. It's just a cross section of Times Square, Broadway, and also the Upper East Side and the Upper West Side

where a lot of elites live, a lot of older, wealthier people. So it's just one of the more prestigious districts, congressional districts, and the entire country. and at the same time, New York City is very exposed to AI job disruption. Brookings called the city the country's most, quote, AI exposed county, a fifth of the workforce do jobs AI could possibly take, we're talking finance and marketing and other white collar jobs like that. So that is probably why there was $20 million spent on this one congressional. It wasn't even the election, just the primary. All right, moving on.

If you want something done right, do it yourself. such is the thinking behind Meta's new AI glasses that came out yesterday and no longer carry Rayban branding. For many years, meta raybans were synonymous with the category of smart glasses. It was a big advantage to piggyback on styles familiar with consumers rather than make your own from scratch. Cough, snap, cough. But Meta is taking the design in-house for one key reason. Price. The new meta-glasses will start at $299, which is less expensive than the Rayban models,

which begin at 379. Meta's not going at it completely alone. They're still partnering with Raybans' parent company, a cell sur, luxottica, on certain aspects of the glasses, but the in-house frames aim to give glasses-curious consumers more options to choose from. Meta also partnered with Kylie Jenner

to release a pair design by her to make sure the glasses are still fashion forward, unlike some competitors. Under the hood, you'll get access to some new features like live language translation and better pedestrian navigation, while also supporting playing music and answering questions about a person's surroundings,

power by META's newest AI model, Muse Spark. Neil, these were pretty well received from a looks perspective, which is always the biggest question mark when it comes to the smart glasses category. Meta thinks they've captured lightning in a bottle here. They, with smart glasses, they built something that people actually want to buy. Shipments of smart glasses were up 167% in the first quarter of 2026,

compared to a year earlier. and meta just absolutely dominates this area. 69% of the market. Their big thing is they want more people to use their AI because they are getting lapped by the other chatbots. According to a Pew Research Survey, 44% of American adults use chat chabit, 24% use Gemini, just 14% use meta AI.

So they just release this new model, Muse Spark, they're going to put it in their glasses. They want people to interact with their chat bot more. and that seems to be the main push for their glasses. And, you know, they feel like they have momentum. And that's why they're going with all these different types of models. It seems like they're trying to break down every barrier someone could have against buying a pair of smart glasses. They are releasing more frame styles because they say, listen, not everybody's face looks the same in every type of glasses.

So we need more models so you can pick and choose what you want. They have more colors. They have adjustable nose pads. they're adding wider prescription support for people who need prescription glasses. They are improving battery life. So their bet here is that if we offer a suite of options instead of just one pair of spectacles, for instance, that is going to lead more people to come into this category. And yeah, they do think that they can absolutely dominate it right now.

And they don't even need the crutch, if you want to call it that, of Oakley and Rayban, these established brands because they think they have a good enough handle on what consumers want to build it themselves. The competition is coming, though. Google and Samsung are collaborating on a new pair of AI glasses that are going to launch later this year. We'll have similar functionality. We know that Open AI and Johnny Ive, who is a former Apple designer, are also working on an AI hardware product. The rumors are that that's going to be earbuds.

And then also Apple is, remember Apple? They're developing smart glasses as well. So this field may not be so crowded right now, but it looks like a bunch of the other tech companies that rival meta are also won in because they're seeing the success that meta has. This announcement, though, of smart glasses may have been overshadowed by another initiative that meta is rolling out. This is reported by the New York Times. Zuckerberg has dispatched a small team at his company to create an app that is similar to Polymarket and Kalshi so they want to get in on the prediction market craze.

It's not going to include, at least if this happens, the codename arena, it's not going to use actual money at the outsets. Can you use like a point system like in video games. But this has already gotten people talking again how Mark Zuckerberg is maybe run out of ideas and is using the copy paste feature to borrow from others to create his own app. When I read this headline, I was very confused

because what does meta have to do with prediction markets at all? But then as I read kind of the New York Times article, it makes a lot of sense because prediction markets have a lot of the same characteristics that social media platforms have, which are highly engaged users who are constantly coming back to check if they are right. It's competitive. You get rankings and these reputations start to emerge.

It is social because people love debating the outcomes and sharing predictions on prediction markets. And it's also lucrative because you do make money on each transaction. So when you put it in that way and just the fact that meta is in the game of attention, absolutely they would be curious about prediction markets because that's what everyone's doing these days, whether you like it or you don't. All right, we're going to take a quick break and come back with a story. story about Netflix and video games right after this. Toby, let's do some free word association. Financially stressed. Toby. What about unexpected $1,000 medical expense?

Crying, stress, lots of stress. Duck? Afflack. Smart man. Afflack pays claims fast, accurately and fairly. Their plans can even help businesses and employees save tax dollars when payroll deductions are pre-taxed while coming at no direct cost to businesses. To learn more, just head to aflac.com slash morning brew daily. That's a-f-l-ac-com slash morning brew daily. Neil, are you a square?

Well, people have called me one since I was a kid, so... Well, now squares are officially cool because square is the easy way for business owners to take payments, book appointments, man of staff, and keep everything running in one place. Whether you're selling lattes or cutting hair, Square helps you run your business

without running yourself into the ground. And right now, listeners can get up to $200 off Square hardware when you sign up at square.com slash go slash morning brew. That's SQ-U-A-R-E dot com slash go-slash morning brew. Neil, check out my card. This just says thank you. Yes, that's for the dinner you paid for last week, but you should also check out the Verizon

Visa card. It helps eligible Verizon customers get more from everyday spending. You can earn 4% rewards on gas, groceries, and dining. rewards toward Verizon bills or devices with no annual fee and unlimited rewards with no expiration. For a limited time, new cardholders can get up to $150 cashback. Just head to Verizon.com slash morning brew to learn more. That's Verizon.com slash morning brew. The famous adage goes, if you can't beat them, join them. And for Netflix, that means embracing the smartphone. The streamer knows that it's competing with your phone for attention. You've got a

Netflix show up on the TV, but really you're far more focused on composing a bangor. to the group chat. To keep people engaged in its content, Netflix is turning your smartphone into an interactive controller with a new game called Unhinged. It's a horror title from in-house developer Night School Studio with voice work from Zoe Kravitz and Sadie Sink coming to the app June 30th. In the game, which only takes less than an hour to complete, your phone transforms into the lead character's phone, which you'll use as a flashlight to illuminate the darkness and answer text and calls. As Netflix's head of narrative games, Sean Crankel told the New York Times, If you're ever looking at your phone, it's because we want you to.

Netflix is desperately searching for a breakthrough in its video games unit, which launched in 2021, but has struggled to find a wider audience. It's hoping that by jacking up the level of immersion and leaning into people's phone use will present a gateway drug for users to discover its other video games. I think this is so smart on so many levels. One, the fact that the technology is just dead cool. And that's actually what developers said is that they discovered that you can point your smartphone. out of TV and it will reflect the flashlight on the TV. And once they came upon that, they're like,

oh, we can build an entire horror game around this. So the fact that a tech insight led to the development of this, I think, is very cool. I also think the length of the game is very smart because they realize that when do you sit down on the couch to watch Netflix, you're probably on the couch for 20 to 50 minutes. Let's make the video game the same length as that. The play time here, if you're good, it's 20 minutes. If you want to take your time a little bit, it's closer to 50 minutes. that is much more akin to an episode of TV than a traditional video game that sometimes stretches on four hours. And I also love that they leaned into the fear associated with being in your own apartment. So this game basically takes place mostly inside where are you going to watch it for

from, it's going to be inside on your own couch. So horror design for the living room is such an interesting wrinkle here in something that does feel uniquely Netflix in a way that maybe some of their other video games has not thus far. And the big picture here is that, Netflix is kind of on the back foot. Remember, Matt Damon came out and said that he's heard that directors and showrunners have been instructed by Netflix to reiterate plot points three or four times in the dialogue because people are distracted with their phones. A report came out a couple years ago that said just 1% of people played Netflix video games despite this big push. Also, if you look at their stock, we did the stock market story earlier, but Netflix is not doing

well on the stock market. It's down 30% this year. It's down 50%. It's been cutting and half over the past year. It is battling for attention from Instagram, from TikTok, from YouTube. All these factors are playing into just sort of investor souring on Netflix right now. So it's throwing a lot of things at the wall. It's trying a horror video game. And I don't think it's a coincidence that this is, it's a horror because horror is huge right now. Backrooms and obsession are dominating the box office. And they released a new hot one spinoff. So they're trying a lot of different things to get back into Wall Street's good graces. Who would have thought, though, that the video game that may not be popular from Netflix is not tied to one of their existing

franchises because they tried that. They tried to do a black mirror video game. They tried to do a squid games experience on the TV, but this has no relation to anything that they've made before. It's an interesting bet, but I do think that this bet on bite-size interactive entertainment is something that will succeed for the most part. I'm interested in playing it, which is I think their target audience. I don't play video games. I don't. I don't even really watch Netflix, but the fact that I can use my phone and like pointed out as a flashlight on the TV, I think that is very compelling. Well, I also thought this was interesting because they do want people like you who don't necessarily play video games. And to figure out how to hook you,

they look to the early 2000s Nintendo consoles like Wii and Nintendo DS, which I'm, you probably didn't play video games, but I'm guessing you played the Wii. I did play the Wii. So they looked, so this was their inspiration is those early 2000s Nintendo consoles that just hooked people very low barrier to entry. They weren't gamers, but they're like the Wii, that seems like something that would be accessible to me. And that's sort of what they're trying to copy. I was a Wii sports resort legend back there. Ping pong, you know, the plane flying game, archery. People listening are like, this is the least surprising thing I've ever heard. It's very fun. All right, let's switch to the finish with some final headlines. Yesterday, federal

officials announced charges against 455 defendants across 45 states and territories, and what was the largest coordinated healthcare fraud enforcement action in Department of Justice history, and the second biggest fraud takedown ever by dollar amount. The cases range from unnecessary medical procedures to opioid fraud to even a skin graft scheme where tissue was marked up by 2,000 percent and billed to Medicare for patients that didn't even need the treatment. Prosecutors say some defendants use the ill-gotten gains to buy luxury cars, watches, and homes. One funded the construction of a beach resort in the Philippines.

The newsletter 1440 wrote that healthcare fraud costs the U.S. between $100 and $170 billion annually, which is roughly 3 to 15% of total health care spending. This was one of the biggest operations ever cracking down on it. Pretty lucrative until it wasn't. Up next, a new audiobook of Homer's Odyssey was released yesterday, narrated by Michael Kane, the legendary British actor you might know from dishing folksy wisdom as Alfred in the Dark Night trilogy. except the voice you hear isn't exactly Michael Keynes, it's an AI clone.

Kane has licensed his voice to the AI audio company 11 Labs, which produced the audiobook using 20 other AI voices, plus AI generated sound effects and a musical score. Given all the hype around The Odyssey this summer, with Christopher Nolan's movie coming out in less than a month, it's going to be a fascinating test case to see whether audiences gravitate toward AI narrated audiobooks. So far, even though AI narration is widespread, just 16% percent. of audiobook listeners said they listened to an AI-voiced audiobook. Also, reviewers did not like the cloned Michael Cain voice. They did say you can definitely recognize it as Michael Cain. He does have a very recognizable voice, but then they were saying that during the biggest moments of the book, it lacked emotional depth. Some scenes that they pointed to when Odysseus comes back and sees all

the suitors occupying his home. I don't know if this is a spoiler alert. If you haven't read the at this point, you got to read it. He shows very little anger when he's recounting that. The frightened suitors sound oddly calm in that situation. So some of the big, you know, emotional moments just fall flat when you don't have an actual person at the reins there. One part of the puzzle that definitely likes us are the actors themselves. I mean, Michael Kane is obviously very much on the tail end of his career.

The fact that he can license his voice and still make money off of his voice is something that is appealing to actors. I don't think this is the last of the AI enhance creative, you know, workers debate that we are seeing because there are two parties that are on other sides of the debate, honestly. It probably will get better eventually,

but I'm not sure it's quite at the level that consumers are going to want to hear a pretty flat, emotionally listless voice reading an all-time classic. I wish he's still alive. I wish we just got him because he does have one of the best voices of all time. Either way, with the Odyssey coming,

you're probably thinking, I should probably read this or listen to it. You are very much a proponent of listening to it, whether it's the Michael Cain AI generated version or any other audiobook of The Odyssey, because this is the way it was originally intended to be consumed back 3,000 years ago. Long oral tradition, baby. That's what I kept saying when I would say, yeah, I'm reading the Odyssey. Well, I'm actually listening to The Odyssey, but you know what? That's totally okay in my book. Finally, we close out every Wednesday show with Suggestion Box, where Toby and I each share a recommendation to get you over the hump of the week. I'll go first.

We didn't play a pre-show game, but I'm just calling it right now. My rec is to gamify your walk. So we all know walking is important and it's especially pleasant in the summer, but sometimes you just want a little something to spice it up. Well, Rob Walker, the author of Art of Noticing,

has a really fun idea. Just start counting random things you expect to encounter on your walk and don't end your walk until you hit that quota. For example, maybe tomorrow I'll go for a walk around the city and keep going until I see 20 people in U.S. soccer jerseys. You could also count dogs, signs, honks, people wearing hats, really,

anything to occupy your mind and stop it from pondering all the things you need to take care of. I absolutely do this on runs more than walks, but I have come back actually from runs on the West Side Highway and quiz you. I'm like, what sport do you think I saw the most of of jerseys that people were wearing? And it was always lacrosse. Like, how many Ivy League schools do you think I saw? But it is fun. I mean, all you're trying to do is pass the time.

If you're not listening to The Odyssey on your walks or runs, maybe do this. I think that is a great recommendation. It's also good on road trips where you can take bets with people in your car and you'll say, like, how many subways are we going to see in our car until we get to our destination? So it's a good game to play as well. My recommendation is the website absurdly optimized.com. As the name suggests, the site's owner Ben goes to absurd lengths to optimize things like pancakes or French fries or nail clippers. For instance, on the absurdly optimized pancake page, the description is a systematic investigation of

acid-based neutralization, CO2 production kinetics, gluing inhibition, and the my yard reaction as applied to a 125 gram flour batter with an interactive stoichiometric calculator that adapts to whatever is in your refrigerator. You can then go down on the page and move little sliders that adjusts the recipe based on tang and fluffiness. The French fry page is fantastic. He also does product reviews of things like nail clippers and hand mixers. I just love people whose brains work like this where he made a subpar bats of french fries. And instead of just stopping there, he goes, you know, no, no, I'm going to go down the rabbit hall and figure out how to make good french fries.

You see the problem. You put it in the work and you solve the problem, even if it's as small as a pancake. We may have to retire this segment because I just went to this website and it has literally every recommendation for anything in your life. I'm seeing absurdly optimized meat slabs, making your bedroom. steamier, using science to plunge your toilet, towel absorbency, quantified. So there's literally everything. I'm not sure what we could add beyond what this guy has at absurdly optimized.com. He covers all the bases. He genuinely is just the best of humanity, I think. I genuinely think when you

make a pancake, most people just put the biscuit in and you're done. But he looked at that and goes, I know there's more variables here. What if I tested every single one of them and brought it to the masses? It is a usable website. Like, I encourage you to go make. a recipe based off of what he has optimized. So it's very useful. It's very fun. And shout out. It's certainly optimized. All right. That is all the time we have. Thanks so much for starting your morning with us and have a wonderful Wednesday. To share your thoughts on the episode or anything else, send an email to Morningbrewdaily at morningbrew.com or DM us on Instagram at MB Daily Show. Let's roll the credits. Emily Milliron is our supervising producer. Raymond Loo is

our senior producer. Our producer is Olivia Graham and our associate producer is Olivia Lake. Technical direction by Nina Miller. Hair and makeup is taking a walk. Couldn't you have done it at any other time. Devin Emery is our president and our show is a production of Morning Brew. Great. So today, Neil, now stick around for our conversation with meta-cTO, Andrew Bosworth. Andrew, thanks so much for joining us. Thanks for having me. Big day for you guys. So my first question is, I wear glasses. They help me see better, but don't do a whole lot more than that. Still, life's pretty good. I manage. Why would someone like me switch over to meta glasses rather than just buy another standard pair? Yeah, I mean, if you're already wearing glasses especially, now you've got audio, you've got video camera capabilities, and you've got AI.

And honestly, if it was just even the first two, it's enough. The most popular thing in the glasses is people listening to music, being hands-free podcasts, by the way. Great on the- Immediate plug, we love that. They play great on the headset. You can do calls, the sound, the audio is great. Super popular in particular with people who are, you know, biking, high-es-es-es-es-es-old. going out for runs. And then you've got the hands-free capture. So you're out there. The best camera is the one that you have with you. Don't worry about taking your phone to your pocket. You've got

hands-free on your head. For me, especially as a parent, young kids, 11-9. As soon as I get the camera up, first of all, I feel bad. It's kind of between me and the moment. But it doesn't matter because 11-9, they're gone. They see that camera coming out a mile away. They are not doing whatever cute thing they were doing before. So it allows me kind of do that, share that with my wife while still being present in the moment. That's enough. Then you've got the AI features, which really have just gotten so much better in very recent history that, you know, this was something was kind of a fringe benefit even six months ago in the last couple weeks. Now it's major, major use case for the glasses. And you can use it conventionally. You can ask

things. That makes sense. But you can also control the device with it. So your hands free, you're on a run. You're on your bike. Hey, take a video, you know, take a photo and send it to my wife, that kind of thing. So it really is, I think for somebody who wears prescriptions, it's kind of a no because you've got to add all this extra functionality. For me, I don't wear prescriptions. I still like to wear the glasses. So it's a good sign. But so you mentioned the phone multiple times

and kind of describing some of those use cases. Are these meant to replace the phone now or maybe eventually? I think over the long period, we can imagine that. Your phone's awesome. Your phone has all kinds. It's why you keep all your stuff. It's great.

We love it. When we started this project years ago, our thought was, what are the things that are so valuable if you could keep your phone in your pocket and still do them? So right now they work with your phone, It's in conjunction with your phone, whether it's Android, whether it's iOS, doesn't matter.

They help. What's cool is, yeah, as we start to see this technology progress, you can see more and more ways to keep your phone in your pocket and still take advantage of the functionality that's there. And over the fullness of time, you could imagine not just glasses, but a wearables kind of constellation of devices that you have replacing that functionality. But that's not where we are today, and it's not going to be where we are for a while. So this is really something that you use in addition to your phone. But so when I go out of the house, I do the phone wallet keys check.

Right. Classic. I feel naked if I don't have my phone. Is there a feature that you can point to now or maybe eventually where it will start being like phone, wallet keys, glasses? And I feel naked without my glasses out there. Yeah. I mean, the way that this is headed with AI especially is if you think about AI today, a lot of

the work that you're doing with AI is just giving it the context that you need for it to answer the question. And so you're like typing in, okay, my wife. if I had this conversation, we're trying to make this plan for next week. What do you recommend for these types of things? I think there gets to a point where if you have AI that's able to be fully contextually aware with you. So it sees what you're seeing. Here's what you're hearing. It's private. It's just for the AI. It's just to help you. Now you can ask any question you want, and it's already on the spot with recommendations if it hasn't already prepared it in advance.

I think it becomes kind of an unmissable accessory in that AI future. We're not the there yet. We've got a ways to go. You can do it right now. You can activate a live AI session, and so you can have it be live for a period of time, but your battery will die in about an hour. So it's kind of a tight tradeoff. So we've got work to do before we get to that point. For me, at least today, though, we literally started listening. What are the top reasons people get their phone out of their pocket? And like music was number one. So it's like, okay, number one, got to do music. It's like, calls is number two. Great. Camera's number three. Great. We've got to do that. So that is how we're approaching.

Wait, did you actually do research on this? Or you just kind of like, No, we absolutely, we did research. We actually, it's like, what are the top 10 use cases that people take their phones out of their pockets for? Can you do those? And so when we added, you know, wear display glasses, it's like, cool, messages, you can do those there. So all these different pieces that we're constantly thinking about, how do we continue to bring the value that you get from your phone to a more convenient form factor? Okay, let's talk about the form factor. So one phrase we keep hearing from meta and all the rivals is that glasses are perhaps the most exciting hardware category of the AI era.

why glasses, why isn't it phones or any other form factor, like a pin or a pendant or things like that? Yeah, there's two answers to that. The first one is glasses are already socially acceptable. It's a hugely, you know, widely worn, people wear them all the time. You're used to seeing them, you're used to seeing them indoors, you're seeing the outdoors. We have a style and a language that we've evolved over a long period of time to deliver these things in ways that are attractive. And so it's just comfortable for people to wear glasses. Do I think it could also be headphones and pendants? And sure, I absolutely think people don't all want to wear glasses. It's totally totally fine. We should have lots of different ways of doing it, but glasses just are a thing

that a lot of the population already wears and all the population is comfortable with. And the second reason is to what I said before, AI, what it wants most of all is the ability to have the context to answer the question. And these see what you see, they hear what you hear. And then you want to have some system for getting that context back into your brain. And that's either going to be a display, which is in front of your eyes, or through audio, which is kind of by your ears. So they're a natural place hanging out right by our major input and output devices. our eyes and ears. In the meta lab in Northern California,

are you guys working on anything else besides glasses that may deliver AI to people's bodies? Oh, yes, we're working on all the things. Vyriada Labs has a very expansive, investigative portfolio of devices that we think could be great. And we're constantly building things and then making a decision relatively late in the build,

like, is it good enough yet? Should we ship this or not? Or do we want to try again? And yeah, we're quite proud of that. We've been doing this for a long time. It's very exciting for us to be at the point now where we're getting pretty good mass adoption of the glasses. And what we're launching today really is just more styles. And because, frankly, get rid of all the features, if they don't look good and if they're not comfortable, people aren't going to wear them.

Most consumer electronics want everyone to adopt the one same style. We've never believed that. We've always believed this is going to have to invert that. And you're going to have to meet every person where they are. And so it's not enough to just do glasses. You have to do all these form factors. It's not enough to do one style of glasses. You have to really support a broad range of glasses.

So, yeah, the portfolio that we have to manage at reality labs is pretty expansive. So I want to drill in on that fashion aspect because I was watching Zuck's Instagram video where he posted about the glasses and he said it's almost fashion first and technology second. You're kind of responsible for that technology aspect. How has it been trying to manage the fashion aspect while also saying, hey, we need battery life in these things. We need to actually work. Now, what makes you think I'm not? No, just good. The fashion guy. No, I think for us, this is a thing that we saw, to our great credit, I think, we saw early on, hey, we need to be experts here. We're not going to be experts here. Let's partner with Esaulexotica. Let's partner with the people who are the best in the world at this. And so, obviously we have the Rayban meta, we have the Oakley meta glasses. These are meta-glasses. They're still made with our partnership with Esaulexotica. They're still our manufacturing partners. They still make the lenses. They're still doing distribution for us. So these really are still made with, in conjunction with,

the best glasses company in the world. So we are partnering with, I think, the best in the business when it comes to thinking through how to position these in the market. You also collaborated with Kylie Jenner on one of the models. Did you interact personally with her? How was like the collaboration effort between someone who's clearly very fashionally knowledgeable but maybe isn't as included on the technology side? Yeah, I want to give a shout out. It was awesome working with her. So our industrial design team spent a whole day with her early on and then again later in the development cycle. And this isn't the case where it's like, oh, let's like get some money and put some name on it. No, no. Kylie was involved in the design and the development. Like everything from

the shape, the materials, the finish, the sizing, to the case and how it was going to be positioned in the market to the campaign that's out there. If you've seen the billboards, they're so cool. No, it's actually tremendous shout out. Because, you know, as you can imagine, people who got computer science degrees don't predict themselves in some future to be in a world where like they're working with Kylie Jenner on fashionable products. No, I was super, we were super blown away. She was she had really strong opinions about the design. She really put her stamp on it. I think it was a great product and I think you can see it in the result. Some of the biggest question here for a lot of consumers is privacy. People do not like to be filmed in public unless they're on the streets after

Nick's game. How are you guys alleviating the privacy fears? Yeah, I mean, I think we've got a couple really things that we're proud of. Obviously, even from the first generation, we've put a light on, so when you're recording, which your phones don't have, right? And I think there's, you know, quite a few videos of people in the subway in New York being surreptish as they recorded. And we've all kind of adapted as a society to the fact that phones are these devices that are capable of video recording at all times. But at least there is a phone out. So you know it's possible. We decided to be a little bit more overt than phones with our glasses by putting the light up. And then in the second generation, we actually added like tamper-proof work. So like if the light is

obviously tampered with, then the video won't record. There's a cat and mouse game happening there and you kind of constantly have to kind of iterate against bad actors. But for the most part, like we are trying to be very forward because the truth that's not just for the people around you, it's for you. You want to wear the glasses because they make you look good. They have, you know, they're useful to you, but you won't wear them if they make other people around you feel uncomfortable, right? And so it's really important for everybody involved that these be something that everyone feels like, yep, this is on the up and up. We understand how this works. This is okay. So I've tried the Oakley meta glasses because I'm a runner and I have run down the west side highway

with the indicator light on that I'm filming. And I get some, I get some looks for sure. How have you manage sort of almost the stigma that comes with that light being on? Even though people are aware of it, they're still not, they're still a little wary of being it. Yeah, we're still in like a very norming phase. And I think I do to go back to the beginning of cell phones when cell phones first had cameras and the degree of anxiety people had that time. This is even before the iPhone. I mean, when Flipphones first had cameras and people were trying to figure that out. So I think there is like a period where we as a society kind of start to build norms around these things. I will say if you look at the popularity of the product, like the numbers that we're selling

and the much of the amount that they're being used, it does seem quite out there. Like it's out there. It's in the world. Every time I go to a kid's sporting event, you're seeing it. Actually, they've been hugely popular with the World Cup because they also do live translations. So people who are traveling or getting translations. So my sense is like we are now in the where these are becoming normal and people are like, yep, I see them, I know what they are, I understand what they are. But that's not a complete process. That's something that's going to take a while, as it did with phones, as it has with any technology that kind of suddenly gains penetration in the market. So I was thinking about big technological shifts over time and I came

back to cars, which, you know, they replaced the horse and buggy, but they also led to suburbs and drive-thrus and motels and all these second-order effects. What do you see as some of those second and third order effects around like how society is organized once these become more mass adopted. Yeah, it's funny. I think there are, when I think about immersive technologies, I could imagine things like VR really changing some structures in terms of how we think about movies or remote work even over time. I feel less different. I feel like these are much more integrated into the way that society works today. I think they're not so much going to be displacing physical infrastructure. I think you have

people who are like probably on their phones a little bit less, having a little more confidence that the digital artifacts that matter to them will be brought to their attention over time by AI. I think AI certainly has a powerful effect of its own, separate from the wearables. I think the main advantage of wearables is the assistive opportunity they give every person, like every single person. And we sometimes see this most obviously in people who have disabilities.

So these are tremendously popular with the blind community. We just donated $108,000 pairs to every blind veteran in America last week. The blind community who previously would have been, we have one of our blind PMs told me, she's like, look, I can get to the restaurant. I've got Google Maps and I've got a cane and I can get there. Then Google Maps tell me, congratulations, you're at the restaurant. Where's the door?

So she uses the AI to ask the question, meaningful. But the truth is, all of us at some point are going to have trouble with our vision. We're going to have trouble with our hearing. Like, we have conversation focused on these glasses. We can do live captions on the display glasses. At some point, that'll benefit us. That'll benefit our parents. Memory.

I talked to Oscar Mike, this tremendous veterans organization that works with people who had traumatic brain injuries. They struggle with memory and cognition. They need to know if they took their pill or not. It could be a matter of life and death, tremendously important stuff. We're all going to struggle with our memory at some point. This is going to be all of us. At some point, it's going to hit all of us.

Even without a traumatic brain injury to blame it on, I often can't remember people's names. Like even without that, right? What's his name? Exactly right. Good old What's his name over here. So my point being, like, I think we all benefit from these superpowers. So I think really the glasses are about giving each person

whether they have disability or not, maximum capability of their own facilities to maximize who they are and what they want to do in the world. That's what we mean.

We talk about personal superintelligence, by the way. It's like maximizing you for who you are. Okay, you weren't the only player entering this space. Many different companies

are racing to build AI hardware. Is there a competitor that worries you the most and why is it Snap? You know, I tell my team, it's a true story, and you can ask my team someday. I tell my team,

in our industry, in the hardware space, very different than the software space, every time a competitor makes different choices, for example, on weight and comfort, that's the opportunity for me to learn because it's very expensive for me to put a product in the market, and I can't do 10 of them. So if somebody makes different choices on me than me, and we see how the market responds to that, that's a gift. I get to learn from them on what to do or what not to do. And so I am excited to have competitors in the space give me free looks.

It's like being on the golf green, your buddy, you know, he gets up just right behind you and you're getting a free look on the read. That's amazing to me. That doesn't mean we are paranoid, you know, we're from the Valley, so it's Andy Groves Valley and only the paranoid survive. Obviously, you take every single competitor very seriously and assume that they're completely competent in all their execution. Everything's going to be perfect.

And you work against that as a metric, so we're paranoid about it. But I do like that there's a lot of people investing in the space. I do like that there's enthusiasm for the space and that we get to learn from all these competitors. When a new product comes out, do you just go buy it and order it and put them on and wear them like a normal consumer? Yeah, I mean, not everyone, but quite a few of them, I think, especially if there's something, I think that there's new, especially if there's innovations in how the usability is and the interface design, you know, I want to make sure I'm really on the cutting edge of what's possible and I want to make sure I understand where the alternatives are. What was the biggest engineering challenge in building these glasses? Like when you made these glasses, I know with support from SLO-L Luxetica, but you made them on your own now.

You know, did you think to yourself, wow, making glasses is easier than I thought or harder than I thought? And like, what kind of surprise? It's harder than we thought because we're just used to having a lot of thermal space and a lot of compute, and these are very constrained. So the amount of battery you can put in here and have it be a comfortable weight, the amount of processing you can do on the device, before you have to run the radio, and radio kills your battery. Like, the tradeoffs are very tight. And in a way, this is, you know, engineers will make fun of the embedded systems engineers have done this for a long time. We've just been very generously gifted these tremendous amounts of compute over the course of my lifetime, right? you've just been writing Moore's law.

And laptops and phones and servers are incredible. And when you're going to a very, very small system on a chip, you're going to a very small footprint of memory, very expensive to run radio, very tight battery. Yeah, you're running into some very old school engineering problems that, frankly, my generation didn't have the experience of the previous generation of engineers at getting good at. So we're kind of having to discover those old techniques again

to work really well in these embedded systems that are really a small form factor. I want to zoom out a little bit because we've talked about the glasses, but Meta is a very large organization at this point. And our co-worker here at Morning Brew, Dan Tumie, made a video titled, What Does Meta Actually Do Now? That video got 1.6 million views. So we have you here. What does Meta actually do? Yeah. I think it's one of those things I think happens a lot with communication and media companies is you just become part of the tapestry of society. It kind of reminds me always of that Bill Burr bit of like everything's amazing and nobody's happy. It's like, you know,

someone, it's like we've integrated communication tools like WhatsApp and Messenger and expression tools like Instagram and Facebook so deeply into the fabric of our lives that it doesn't even occur to us that those things require like real work and effort to make and produce and to grow and to continue to make great. And there's other things that have always been hard people to see. The degree to which meta is the engine of small business growth, not just in America, but around the world, undeniable. We talk to any small business owner trying to find their customers. We are the way that they do it. We are driving their businesses. And that's the majority of new jobs created in America in the last decade are coming from small and medium businesses powered by

our technology. I promise you that. And it's a story that's, we haven't done a great job telling, but it's a real story. And it's a story of a global story. It's an American story. And that continues. We continue to be a tremendous powerhouse in the space of connecting people with their audiences. And I've always believed from the beginning, when you do this really well, everybody benefits. The people who are running these small businesses are benefiting because they're finding the right customer. And the customer is so pumped that they found this business. And so that continues to be the major engine for us, even as we also do this work in reality labs, do this work in the meta-superintelligence labs, like to advance these frontier technologies.

We've got to talk about the vibes in Menlo Park. Last week, you reportedly said that morale at the company was the lowest it's ever been given layoffs and a reorg. What exactly broke down and what is your plan to win back employees? Yeah. Well, there's kind of three layers of it, perfect storm as it often does happen. One of which is the entire industry is going through is a period of absolutely insane change, right? This is by far the most insane period of technological change in my career. And I've heard the same thing from people whose careers work twice as long as mine. So this is totally insane change, not just the magnitude of it, but the speed of it. It's happening very quickly. And so there's a tremendous amount of uncertainty for all of us, the core

value prop that I thought I had all else in the world. If everything else fails, at least I can write code, and that's a rare and valuable skill. And that is under some threat. So there's this total, almost industry-wide context of uncertainty and unrest. Seeing this, you know, I think Mark made really bold moves inside the company to make sure that we were positioned to be at the forefront of this shift of AI. And I really, I've been super impressed by it. I'll give Mark a lot of hard time. So this is happening to the thing.

I happen to have been super impressed by. But we did not do nearly enough work to connect those two dots for the company. Be like, look, we're all feeling this anxiety. We understand the change. We're doing this to make sure that we stay ahead of the change. And then you have a third piece,

which is, you know, you're doing a layoff. And a layoff is going to be unpopular. It's going to be hard. It's going to cause people to feel anxiety, you feel an acute uncertainty about what's going on in the world. And so there's like all these different layers. I do feel a little bit fat. I wrote kind of quite a long memo internally to the company of all the different things

that I thought, you know, we could have communicated better or that we did wrong and how we can do it better. And only one of like a thousand points was snacks, but that was the only thing people attached to you. But by the way, it was also the most popular thing I wrote. So it's not a bad thing. But yeah. So, you know, we really are as a leadership team spending all of our time right now trying to talk to employees and be like, look, we screwed this up. We're going to take these steps. Here's the kind of commitments that we're going to make. Here's the stuff that you want to hear from us. We're here for that. I honestly just

my macro thought is like, there's this anxiety. And we were not doing enough to keep them close to us as we made the shifts. We need to make to stay on top of it. It's not just a meta issue, though. AI in general is deeply unpopular just with the general public. According to a recent survey, the only thing more unpopular than AI right now were the Iran War and the Democratic Party, what are tech companies doing wrong when it comes to marketing AI? Yeah, it's a story as old as Silicon Valley, unfortunately.

We call it the hype cycle, right? So the Silicon Valley, almost every technology that comes out of Silicon Valley has the hype cycle. We kind of talk it up to the moon, and then it inevitably fails to immediately materialize all those expectations. So you have what we call the trough of discontentment, and then we like grindingly get it back.

And so I think AI is a very real technology. It's a great technology. It's important. It really will do things like cure diseases and make your everyday life easier. It really will. We haven't actually done all that work yet. So we have this very cool and intelligent thing for some of us who are willing and technically capable of like going through unbelievable complicated morass. We can unlock this power. So we're feeling super in awe of it. But bringing this to billions of people, that is the work. And it is work that takes time, not just to get the process. but also to get people educated enough to use it, to take advantage of it, to care about it. It's like the internet was great. I remember being on the internet, right, like 1995, and be like, this is going to be amazing. The internet didn't really hit mainstream for like another seven, six, seven years, right? Like it just takes time. And so I think the answer to this is like, we are hyping a thing, the costs of which are real

and being felt today, the benefits of which are still theoretical for most people. And it's our fault. Like we shouldn't do that. What we should do is lead with real value. I do think this is the thing, again, I'm very pleased with the focus at Facebook on consumers, bringing this personal superintelligence to every person. Just making your life better.

And you can do that with the glasses. You can do it the meta AI app. It's really good. New Spark, our new model is really, really good. It's great at local. If you want to find recommendations near you, it's great at commerce. You want to help shopping.

It's really good at health. If you want to get some extra health tips on how to better engage your primary care providers. Like, it's really good at several things. And like, this is real value, and we don't have to do it. dress it up. We don't have to oversell it. You can take it or you can leave it, but it's real value. But again, maybe something that's different than comparing this to, like, the internet

revolution is like a lot of AI leaders say, this thing could kill us. Like, this thing could be attained super intelligence and, like, wipe out humanity. How is that, like that hasn't happened in other technological cycles. So this does feel a little different. Yeah, I mean, I would say that there's always somebody who's doing hyperbole, like crypto, there was definitely crypto people who were saying that this thing was going to be, like, all your money is. worth nothing. Like, I do think there exists in every one of these hype cycles. Somebody who sees a line that looks like this and assumes it goes to infinity and is wrong. And then there's a hype cycle crashes. And I think they get like a lot of air time because it's like they're saying

a crazy thing. I don't believe that. I think this is a great and useful tool. I find myself sometimes with no friends in the AI space because I really believe so strongly that AI is an important, profound transformational tool, maybe the most transformational tool since computing itself. Certainly a relationship there where it's making it, making the computers much more useful to us by better understanding us in our context and being better able to help us with what we need. I think making software, something that isn't exquisite and rare, but something that everyone can contribute to is a huge benefit to society.

I really believe in the power of it. And I think in five or ten years, we will feel about AI the way we feel about phones. Like, yeah, this is an awesome tool, and I'm like, my life is better because I have this tool. What I would like for my, at the same time is like, we need this show, not tell. We need to be like, let me just demonstrate you the value. If I thought I was building a thing that was like going to devastate humanity, I would not build the thing. I'm like, I just call me crazy. I simply would not build it.

That's not what I'm building. And we are building AI the same levels as anybody else is building AI. And I believe very strongly, and it's potential to help people. And I don't think it's that kind of thing. So I went to your website, boss.com. And the first thing that, it's a strong domain. It is. Strong domain name.

The first thing that I see is this write up, and you write in one of them. Shame is a powerful force in our society and prevents us from sharing our lessons. We protect ourselves from criticism, but also prevent people from connecting with us meaningfully. How can people overcome shame? Bray Brown, I mean, I hate to be so cliche on the podcast circuit, but Renee Brown really was somebody who I, I did not have even a language for thinking this through until I listened to her research about shame and the impact that it has on people. And I kind of grew up, I would say, with a little bit of that construct, maybe a masculine construct of like project strength all the time, you know, like don't let anyone see any kind of weakness. And it was a very isolating way to be as a young man.

And I lost a lot of valuable relationships doing that. and the craziest thing that happened was I had great friends who were so generous for their time to come to me and like kind of talk to me at the impact it was having on them on people around me and my wife now my girlfriend of the time I got some feedback at work about this and I came home and I was kind of complaining about it and she was like yeah but they're right and she was like saying it with all love and her heart like it was no judgment on me there was no shame she's like yeah but they have a point right like that you do kind of do that thing and it was really shook me to my core and so then I started this complete 180 I was like it was like I was like I was like yeah but I was like you know I was like that you kind of Like, what if I just, like, said everything I was afraid of? And I was like, hey, I kind of think I screwed this up. Like, I did this thing wrong. And it was unbelievable the response I got, the generous response I got from people, you know, owning a mistake like that. And in a way, people perceive me as way stronger, like way more, you know, like, it's like, wow, like, that's real strength. Like, the real strength isn't in like appearing strong.

It's being strong enough to be like, yeah, I screwed that up. That's my bad. And listen, I have time. I have opportunity for this constantly. It's not like I'm not screwing things up at this. I hope to screw them up less often or less badly. But even just this thing with the employees, you know, that we're dealing with internally. You know, I'm out there being like, yeah, we screwed this up.

I screwed this up. Like, we thought you were with us. You weren't with us. We brought you on these changes. We didn't talk to you about them. I know better than that. I should have done better. You know, we're making this a cute change or we're rolling this thing out.

We did a crappy job to explain it. It's not, it doesn't fix the problem, but at least you're able to deal. with it earnestly. You know, I think, and one of the things that's hard being in a big company is you don't have as much of a personal relationship with people, right? And so to them, you're almost a caricature. You're not even a real human with, like, your own foibles and mistakes. And they depend on you. So your mistakes are more costly for them than they are for, like, a smaller thing. So the stakes are very high, and I'm certainly not perfect and continue to make mistakes. And the only commitment I can really make to people is I try to live up to, try to own other mistakes.

and do better the next time. Got one final question for you. You've been with META for 20 years since 2006. Yeah. What keeps you engaged and what lessons can young people who are listening to this just starting out in the workforce can take from maybe your insanely long tenure at a single company? Yeah, it's a shout out really. So there's really two answers to this.

The first one is I work with great people. And everyone, you go back in time, you go back to like great teams of people who've worked together in history. and it is a team game. It's a team story. You know, it's not just that I like working with Mark, but Chris Cox and Javi and Tom Allison and these folks I work with. I've been working with these guys,

most of them, 15, 16, 17 years. My own team, Alex Himmel, got, the guy's been working with me 17 years. And there's a real, it's not just a good vibe, which is nice, but also you can do more. We know each other so well. We can kind of finish each other sentences. We kind of have a sense of, in our heads, I have a version of like, okay, boy, you know who's

amazing at this? Is this person over here? And they're going to come in. They're going to make this work stronger if I go bring them in on this thing. And so the team is a big part of it, truthfully. Even that probably wouldn't be enough. I am such a – I get so bored so easily. Like, I don't – it's like after a year or two of a thing, even before it's done, I'm like,

yeah, yeah, I get it. Like, intellectually, I've figured this out, and I'm, like, not excited about it anymore. I've been very lucky. This has been a company where, like, every two years, I could just do a completely different thing inside the company. If you look at my current job in reality labs, I'm doing robotics now. That's awesome. Two years ago, I wasn't doing robotics.

And so it's like, oh, so now I've an entirely new area to go explore and understand. That's very fortunate. My major advice that I give to young people is to follow the steep part of the curve. Doing the thing that you know you can do well is always a very appealing choice because it feels very safe. and I think you're going to do that and you're going to get bored. And eventually it's going to be the only thing that you can do. And so I advise young people to do is like, find the steepest part of the curve.

Where are you going to learn the most? Early in my career, because I was jumping every 18 months, basically, to new things. I kind of joked, I was like, you know, I was like in karate kid. I was waxing the car. I was painting the fence. I didn't appear to making progress. And my career appeared early on to be moving much more slowly than my peers, who were staying in one place. who were staying in one place and kind of getting promoted up.

And then my career at the end took this more vertical turn because it turns out I knew how to do all the different parts of the value chain from the different jobs that I had done over time. And it's just so much more fun to learn new things. Learning new things is like just amazing. It's what a gift to be able to learn things. And so that's been the real key to my longevity probably is just working with a great team of people

where I look forward to going and seeing them every day. And when I'm challenged, I know that they're going to pick me up and help me out. and make me better, and then having the opportunity in terms of the domain to just keep doing new things and learning more. Hang out with good people, hang out with young people, all good advice. Andrew, thank you so much for sitting down with us and talking through meta and the glasses and everything else. If you want more of Andrew's thoughts and ideas, you can check it out at baws.com, but thanks so much for sitting down with us. Thanks for having me. I'm about to get a lot of referral traffic to this blog.

I think the last thing that was posted was in February. Yeah, I got to have to write a lot of. There's good stuff on there, though. Thank you, thank you. If you're preparing for a podcast, there's good stuff. I'm glad, I'm glad, yeah. The Wired Newsroom is known for award-winning reporting on how technology shapes our world.

On Wired's uncanny Valley, we take that curiosity even further. Each week, journalists from Wired break down the biggest stories in tech while speaking directly with the people building, challenging, and reshaping the future. Is the AI boom sustainable? How do you protect your privacy in an age of constant surveillance? Uncanny Valley tackles the questions driving today's tech debates and lighting up your group chats. Listen to new episodes every Thursday, wherever you get your podcasts.