Prof G Markets - Why Oracle is Crashing Right Now

Episode Date: December 16, 2025

Ed Elson unpacks why Oracle shares are tumbling and what that means for OpenAI with Gil Luria, Head of Technology Research at DA Davidson. Then Helen Toner, Executive Director at Georgetown’s Center for Security and Emerging Technology, joins to discuss Trump’s new AI executive order. Finally, Ed checks back in with AI data center company, Fermi, and sees how the company is doing since its IPO.  Check out our latest Prof G Markets newsletter Follow Prof G Markets on Instagram Follow Ed on Instagram and X Follow Scott on Instagram Learn more about your ad choices. Visit podcastchoices.com/adchoices

Support for the show comes from public.com. You've got your core holdings, some high conviction picks, maybe even a few strategic options at play. So why not switch in the investment platform built for those who take it seriously? Go to public.com slash profg and earn an uncapped 1% bonus when you transfer your portfolio. That's public.com slash prop G paid for by public investing. All investing involves the risk of loss, including loss of principle, brokered services for U.S. listed registered securities. Options and bonds in a self-directed account are offered by Public Investing, Inc., member FINRA, and SIPC, complete disclosures available at public.com slash disclosures.

This episode is brought to you by Square. You're not just running a restaurant, you're building something big, and Square's there for all of it, giving your customers more ways to order, whether that's in-person with Square kiosk or online, instant access to your sales, plus the funding you need to go even bigger. And real-time insights so you know what's working, what's not, and what's next.

Because when you're doing big things, your tools should to. Visit square.ca to get started. Canada's Wonderland is bringing the holiday magic this season with Winterfest on select nights now through January 3rd. Step into a winter wonderland filled with millions of dazzling lights, festive shows, rides, and holiday treats. Plus, Coca-Cola is back with Canada's kindest community. Celebrating Acts of Kindness nationwide, with a chance at 100,000 donation for the winning community and a 2026 holiday caravan stop. Learn more at Canada's Wonderland.com. Today's number 60.

That is the percentage of Americans who say they have had a paranormal experience. These are events that involve visions of ghosts and in some cases demons. In fact, I had one last night when I watched Fox and Friends. Money markets matter. If money is evil, then that building is hell. The show goes up. Welcome to ProfiMarkets. I'm Ed Elson. It is December 16th.

Let's check in on yesterday's Market Vitals. The major indices fell ahead of the jobs report, which will be out this morning. Bitcoin dropped back below $86,000. Broadcom suffered its worst three-day sell-off since 2020. And finally, Tesla closed. at a high for the year after Elon must confirm that the company is testing driverless robotaxies in Austin. Okay, what else is happening?

Oracle stock has taken a serious tumble since last week's earnings missed. The stock fell roughly 11% on Thursday following their courtly report in which revenue fell short and it raised its capex guidance again. Shares then dropped another 5% on Friday after Bloomberg reported that Oracle has delayed completed completion of open a.R. related data centers, and the stock fell another 3% yesterday, as shares remained under pressure. All told, Oracle is off 40% from its September peak. So, here to help us make sense of what is going on with Oracle. We're speaking with Gil Luria, head of technology research at D.A. Davidson, Gill. Good to see you.

Good to see you. Thanks for having me. So Oracle reports earnings on Thursday, misses on record. not a huge miss, but then what we're seeing is a pretty phenomenal reaction down 11% on Thursday, then it keeps on sliding, sliding again after the weekend. What are investors most concerned about right now? Well, Oracle was put in a precarious position by Open AI. Three months ago, Open AI promised them, you know, we're going to spend $300 billion with you over the next five years. And at the time, that was OpenEye's biggest commitment, and it looked like it was going to come to fruition. And at the time, it even looked like Oracle had won that over all the other possible providers of data center capacity.

But what we found out in the subsequent couple of months, that Open Air was making a lot of commitments to a lot of companies, $1.4 trillion in total, which at some point we all realized they didn't actually intend to live up to. And so here's Oracle. They need to borrow a lot of money, to build a lot of data centers, to serve this contract that may or may not materialize. But they have a contract, so they can't say they don't have that. So here they are.

They show up on Thursday and say, we have $523 billion of remaining performance obligations. In spite of the fact that we all know, that $300 billion of that is from one company that may or may not end up spending that. So Oracle is in this tough situation because they already have a leverage balance sheet. They need to borrow more money, and they got on the call on Thursday and said, well, we intend to keep our investment grade rating.

We intend to stay below a 3.5 debt-to-eba-dha ratio, which they are already close to. And we're going to have to find creative solutions to do that. We're going to have vendors own the chips. We're going to have customers own the chips. We're going to postpone leases. So they got themselves into this really uncomfortable position where it's unclear how they can proceed moving forward,

building for a customer that may or may not materialize. Everything you just said there, you know, this relationship with Open AI, the fact that it's the majority of their future revenue is coming from this company. we don't know how much we can trust open AI. We are also looking at the leverage that Oracle is taking on. All of that stuff sounds like stuff we kind of already knew,

or at least that most people generally had an understanding of. But it looks as if something happened or something changed people's minds with this earnings report. Is there something that made people believe all that maybe more strongly or some number that particularly stood out to people? A couple of things. So one thing is that you mentioned that they missed on revenue. And part of the reason they missed on revenue is they didn't live up to their obligations

to build the data centers at the rate they had promised. And that's the premise of a company like Oracle or like CoreWe, these in the old clouds, the whole premise is, oh, you know, Microsoft, Amazon, Google, they can't build it fast enough. Come to us, we'll get it done. So when you have, and this happened to both of these companies over the last couple of weeks and show up and say, actually, I can't get it done in time, then why are you there?

So that's part of it. And the other thing that they had this window on Thursday to own up to what happened to them with OpenAI, to own up to the fact that they'd been played and say, look, there's 223 billion of remaining performance obligations from robust customers that we have a hundred, 100% visibility into, and then there's another 300 billion that we hope to get, and we're working with Open AI, but we have to discount that somehow to be more reasonable. And instead of doing that, they got entrenched in this notion that everything's fine, we can continue to borrow money, and we're going to continue to build data centers, and the customers are going to be there, nothing to see here. I think that in itself caused unease. Yeah. Just looking at their debt situation,

so the long-term debt increased to $116 billion, it's up 44% from a year ago. But I think probably most interesting is what's happening in the credit default swap market where the spreads on Oracle's credit default swaps have hit an all-time high. What does that tell us about Oracle's situation? What does that say about investors' concerns about Oracle right now? That even if they can access the debt markets, the cost of that debt has gone up a lot since last time they raised. Plus, we've now found out because they slipped it into the 10Q on a Friday morning that they have committed to $246 billion of operating leases. Above and beyond the $108 billion of debt, they now have added a colossal amount of operating.

No, mind you, that's over 15 years. but that's still a very big commitment. Add to that, the activity in the CDS market, which tells us that their debt is now worth less than when they issued it, that means that next time they have to raise debt, it's going to be harder and more expensive, which, again, makes it harder to believe

that they can build out the data centers at the rate that they need to to live up to the obligations that they have made. It really seems as though Oracle has become, kind of the proxy for the whole AI bubble or maybe the tracking stock for all of the concerns about AI. Everything you're describing where you had this massive surge right after the Open AI announcement. Everyone gets very excited. Went from 220 to 330 after that deal. We're back down to 185. We're down 40% from that peak, which is just extraordinary. I guess,

my question is, have we gone too far? I mean, if we were at 220, now down to, then yes, we had the rise, but now down to 185, I mean, people are less bullish and less excited about Oracle than they were, even before the Open AI announcement. What do you make of that? Is it possible perhaps that maybe they're overselling? They still have to unwind the situation. Until they get out of this predicament, the stock is going to trade at a discount to their growth rate, which is now actually accelerated and they were actually able to deliver a nice growth and growth expectations for next year. But until they get themselves out of this predicament, which will mean renegotiating with Open AI to something more reasonable, investors are going to be very cautious about engaging with Oracle, not to mention that, again, investors feel misled. Oracle was misled by Open AI, and then Oracle misled investors. There's always a discount when you mislead investors, and that tends to linger for more than a couple of months.

It could linger for a while while investors regain their confidence in Oracle's leadership. Just before we let you go, what does this say about Open AI? I mean, Open AI is a private company. We don't get to see the stock price moving up and down, but I mean, what we're really describing here is a crisis of faith in OpenAI, which bled into Oracle because of their relationship with each other. I mean, does this mean Open AI if it were trading would be down like 50% at this point? I mean, it doesn't seem good for them. Most likely, yes. And really, we're at a very critical juncture right now. What the company has said, what Open AI has said

is by the end of the year, we're going to give you this plan for how we intend to get capital to fund all these endeavors that we've been talking about. Well, we're in the middle of December. Within the next couple of weeks, we're going to find out how much Open AI was able to raise. We know they've been on the road trying to raise capital. If they raise $5 billion or $10 billion, that's not nearly enough to pay for everything they've signed up for. But if they come up, if they show up in a couple of weeks or even as early as tomorrow and tell us, we have $100 billion of capital lined up, equity, debt, government back. Saudis, Gulf states, SoftBank,

then I think there's going to be a big relief in the entire sector. I suspect based on everything we've seen, we may have the first scenario, not the second scenario ahead of us over the next couple of weeks. All right, Gil Lurie, head of technology research at D.A. Davidson. Gil, always good to see you. Thank you.

Thank you. After the break, Trump signs an executive order on AI. If you're enjoying the show, give Profi markets a follow. Support for the show comes from Grooons. The holidays are a time to indulge, but even if you're eating more than you typically do, you might not be getting the nutrients you actually need to end the year on a high note. Grunz may be able to help you fill the nutritional gap so that you can enjoy it all guilt-free. Grunz is a convenient, comprehensive formula packed into a tasty little pack of gummies.

this isn't a multivitamin or green scummy or a prebiotic it's all of those things and then some at a fraction of the price and bonus it tastes great every grown snack pack is filled with six grams of prebiotic fiber which is more than what you get in two cups of broccoli plus grunes are nut gluten and dairy-free vegan include no artificial flavors of colors and are backed by over 35,000 research publications don't let the holiday travel hosting parties and late night set you back give yourself a little extra support so you can enjoy all the holidays magic get up to 52% off with code markets at grunz dot co that's code markets at g r u and s dot ceo support for the show comes from lincoln one of the hardest parts about moving to a new city is finding your people you can look far and wide but it's hard to find the people who just get you And the same goes for B2B marketers. Locating the right people who align with your business and an audience that connects you with your product

and your mission can make all the difference. But instead of spending hours and hours scavenging social media feeds, you can just tap LinkedIn ads to reach the right professionals. According to LinkedIn, they have grown to a network of over one billion professionals making it stand apart from other ad buys. You can target your buyer by job title,

industry company role, seniority skills, and company revenue, giving you all the professionals you need to reach in one place. So you can stop wasting budget on the wrong audience and start targeting the right professionals only on LinkedIn ads. LinkedIn will even give you a $100 credit on your next campaign so you can try it yourself. Just go to LinkedIn.com slash markets. That's LinkedIn.com slash markets. Terms and conditions apply only on LinkedIn ads.

Canada can be a global leader in reducing the harm caused by small. But it requires actionable steps. Now is the time to modernize Canadian laws so that adult smokers have information and access to better alternatives. By doing so, we can create lasting change. If you don't smoke, don't start. If you smoke, quit. If you don't quit, change.

Visit unsmoke.ca. We're back with Proftly Markets. President Trump just signed an executive order aimed at blocking state AI laws this year alone. Every state has introduced AI legislation and 38 states have already passed 100 laws. Now the idea is to replace this patchwork of state laws with one national framework. So what does Trump's order actually do to accomplish that? How far can it go to answer these questions? We are speaking with Helen Toner, executive director at Georgetown Center for Security and Emerald.

emerging technology. Helen, thank you for joining us on Profi Markets. It's a pleasure to be here. So Trump signed this executive order on Thursday. Basically, the idea is we're going to regulate AI, not at the state level, but from Washington. Let's just start with your initial reactions to this move. It's been really interesting to watch the play-by-play here. So this is not the first move. The Trump administration did not do this out of the blue.

Earlier this year, there was an attempt to pass preemption as part of the one big, beautiful bill. So preemption being, you know, the federal government stepping in, saying we have rules, we're going to, the rules we have are going to preempt anything that the states do. So you have to follow the federal law and not state law. They attempt to do that with one big beautiful bill. The trick being that they didn't actually have anything, there were no federal rules, right? It wasn't, you know, here's the rules for how to sell such and such. and now all the state laws don't apply. Instead, it was there's no rules and also states aren't allowed to make rules. And that turned out to not be very popular in the big, beautiful bill,

discussions and end up going down in a Senate vote of 99 to 1. They then took another stab with the NDAA, National Defense Authorization Act. This is a must pass defense funding bill. That also failed in Congress. And while it's unusual to, you know, preempt state law with nothing to replace the state law, they did at least go through Congress in both of those attempts. They were trying to say, let's do this the way that you're supposed to do it. The federal Congress passes the law that overrides the state laws. What the executive order from last week did was really trying to do an end run around Congress instead. So it has several different sections in there, each of which is sort of on its face, something that the executive branch can do,

that the president can, you know, require of government agencies. But that altogether, when you put them together, are trying to do something which the executive branch really can't do, which is pass a law to preempt state laws on AI. So it's really trying to intimidate states. It's trying to threaten legal action against them. But it's all going to come down to messy court fights and, you know, do they get access to federal funding and things like that. You know, when I look at what he's done here, it's hard for me to think that this wasn't a result of the hundreds of millions of dollars of lobbying spend that was coming from. from these AI companies and the fact that David Sacks is in there, who obviously has been reported

frequently is an investor in many of these AI companies. But I just want to push all of that aside because the question of should we regulate AI at the state level or at the federal level, I think is actually a fair question. Absolutely. What do you think? How should we be regulating AI? I think there's no one approach that's going to be enough to regulate AI as a whole because AI is so many different things. So I think we need to be thinking differently about how are you regulating AI in education? How are you regulating AI in the military? How are we regulating AI in all kinds of other places? So a lot, if you look at bills that states have passed, many states have passed bills about how the state government uses AI. That seems like something that you really want to regulate

at the state level, right? Alabama decides how the Alabama government should use AI and so on. Other things like some of the more sort of national security focused issues, maybe there is actually a strong case for that being something the federal government should regulate. But then you see states in this position of, okay, these issues are moving fast. Companies are warning us that these threats might materialize in the next year or in the next, you know, next year or in the next two to three years. And if you can, if you're looking at U.S. Congress in Washington, D.C., it's pretty clear that they're very unlikely to regulate on those issues anytime soon. So what do you do with those issues where it's not actually obvious that the

states are the right place to regulate it, but it is obvious that federally we're not going to regulate? You know, do you then say we should tie the state's hands and they shouldn't be allowed to do anything while D.C. goes ahead and does nothing. And that's, I think, where you have some legitimate disagreement. Personally, I tend to think that, you know, the laboratory of democracy the idea is valuable and is a sort of foundational principle of how the United States works. And so even in those cases, you would want to have the states able to, you know, take a stab at it. I just want to read you a quote from Trump. He said, following this executive order, he said, we have to be unified. He said, quote, China has one vote because they have one vote,

and that's President Xi. He says, do it, and that's the end of that. Basically saying, that's a good system. I just want to get your reactions to that quote because, you know, in a lot of ways, you could argue he's right, and that is China has the ability to just make things happen. The executive order comes in.

He has basically total 100% executive authority, which allows them to get stuff done quickly and pretty aggressively. We have a system of democracy which slows things down. He's basically saying he doesn't like that system? What are your reactions to that quote? I mean, I have to prevent myself from getting sidetracked. I'm somewhat of a scholar of Chinese science and technology and their regulation. And,

you know, my mind immediately goes to, well, actually, you have the cyberspace administration of China and there are kind of, you know, there's tensions with the Ministry of Science and Technology, and then the companies are kind of pushing back in the, you know, so in China, I think it's not quite that simple. Fair enough. But, of course, it is an authoritarian system, and Xi Jinping does wield an enormous amount of power. So point taken. Yeah, I mean, I think it's just fascinating to see this from the leader of the party that for so long has protected states' rights and has understood the importance of that and has fought for it in so many other settings. And, I mean, you are seeing plenty of Republicans reacting to both the, you know, the two initial pushes in Congress to preempt state AI laws and also this executive order. They are pushing back on it.

They are defending states' rights. but, of course, the CEO has now gone through and the federal government is going to move to execute on it. So it's not clear to me what Republicans in Congress can do about that at this point. Just before you let you go here, I mean, the politics of AI, this is increasingly becoming a political issue,

how we regulate AI, who stands to benefit from AI, etc. And what we continually see in the data and the polling is that Americans do not like AI, increasingly so. And so it does seem striking that Trump, who does present himself as something of a populist, he wants to go with whatever the most popular answer is, is seemingly taking the other side of this. He seems to be siding with Silicon Valley in a lot of ways,

and he wants to let AI run kind of unregulated, unregulated. I just interested to get your reactions on how this might play out over the next year or so. And does this executive order basically just mean he's siding with the tech bros? And how will the nation react to that? I think it's the right question. I think it's really striking at whatever polling you look at how negative public sentiment about AI is. In some ways, I think too negative. People are very negative on self-driving cars, for example, which I think an incredible technology when you look at the safety data and how many lives from traffic deaths could be saved. But I think the tech companies really need to get realistic with

themselves about where the public is at on this technology and how do you, you know, I see tremendous potential in AI. The tech companies see it too. How do you start to deliver on that and show people a positive vision? I think banning state legislatures from creating any rules or, you know, putting any safety guardrails on is not how you create that confidence. So I definitely think there's a chance that we'll see significant backlash to AI over the next few years, and this certainly isn't going to help prevent that. Helen Toner, Executive Director at Georgetown Center for Security and Emerging Technology. Helen, thank you very much for joining us.

Appreciate your time. Thanks, Ed. Before we end, we wanted to give you a quick update on a stock that we highlighted a couple of months ago. You might remember the company Fermi, which is an AI Data Center company, that IPOed in October at $19 billion. And you also might remember what we said about Fermi. We said that it was a, quote, shit show of a company. We pointed out that it would likely be the, quote,

worst IPO of the year. And we said that ultimately the stock would come crashing down. This is not going to end well, and I will lock that prediction in right now. Ultimately, as we have seen time after time, gravity will hit. Gravity must hit. And this thing will have to come down. Well, here is our update. The crash came sooner than we thought. As of today, Fermi stock is down almost 75% in just a couple months since it went public. The company is

now worth around $5.5 billion. And the reason the stock collapsed is really because of what we said, and that is the revenue that it had promised, is not really showing up. More specifically, last week, one of their first customers, a company that had planned to rent compute from Fermi for $150 million, that customer pulled out of their purchase agreement. Supposedly, there was some disagreement or a fight over the terms, and it led to just a termination of the entire relationship. Now, to be fair to Fermi, $150 million actually isn't all that much. I mean, they burned about $350 million last. quarter alone, they would argue that they're, you know, thinking bigger than these numbers.

The problem isn't the number, though. The problem is what it says. It's the fact that one of the only contracts for a company that hasn't proven itself in any capacity, the fact that that has fallen through just a few months after they debuted, that is a sign. And it is a sign, as we proposed on the show, that perhaps these guys don't really know. what they're doing. Perhaps they saw this buzz around AI and data centers and nuclear and they went out. They built a fancy deck. They made some spreadsheets. Rick Perry came along and they raised a bunch of money. But that tells you nothing about their ability to actually execute and generate business to actually turn these promises into reality. And what we know about this company is that they

haven't executed anything. They haven't built anything. They haven't signed a single thing. They haven't tenant. They have no profits. And most importantly, as we highlighted, they still have no revenue. So it appears that Wall Street is beginning to come around on these issues. The stock's down nearly 75%. We predicted this would be something of a disaster. If you want our full analysis, I would encourage you to go check out our episode from October 1st, where we lay out all of the issues with the company. But it does appear that the disaster is now playing out the only wrinkle it happened a lot sooner than we thought. Okay, that's it for today.

This episode was produced by Claire Miller, edited by Joel Patterson, and engineered by Benjamin Spencer. Our associate producer is Alison Weiss. Our research team is Dan Shalahn, Isabella Kinsel, Chris O'Donoghue, and Mia Silverio, and our technical director is Drew Burroughs. Thank you for listening to Profti Markets from Profugee Media.

If you liked what you heard, give us a follow. I'm Ed Elson. I will see you. tomorrow.