Google’s Gemini puts OpenAI on shakier ground, with big market implications
Almost every night, for almost a decade, I got a phone call between 7:00 and 7:01 p.m. ET. I didn’t have to look at the three letters on my phone screen to know who was ringing. It was the old man we called Pop, or more like “The Old Man of the Mountain,” as he called himself when we had our grandchildren. Sometimes I tired of the words, but I always took a breath before I hit hello, lest he hear the fatigue in my voice for something I know I would miss dearly one day. “Jamesy,” he would say, “the best one yet.” Always, “the best one yet.” If I have a regret, it’s that I never tape recorded it because I would like to play it between 7:00 and 7:01 p.m. now, every night. But I didn’t. So, call me intrigued, when I saw on my schedule that I would soon be interviewed by two gentlemen, Jack Crivici- Kramer and Nick Martell, on a podcast called “TBOY.” I knew these two as the people who started what I know to be Robinhood Snacks, something I still read midmorning, which is about 6:30 a.m. for the collective slackers I deal with. I had heard of some of their stuff since, but candidly, I didn’t pay close attention — or, at least, close enough attention until I knew I would be interviewed by them on “TBOY.” I have always felt kindred to anyone younger who loves the markets, so I figured this one, this interview, would be the one where they would have actually read my new book, “How to Make Money in Any Market,” and even realize that I was trying to radicalize the public into thinking they could pick a few stocks — five, to be sure — to go with the omnipresent index funds that we are required to take, along with our mumps, diphtheria, whooping cough, chicken pox and measles shots. At a time when so much is up for debate, I have a right to argue that you can buy stocks of companies that you can observe. You know, be curious about them, Google them, look at their websites and discover everything that, in many cases, granted them admission to the sainted S & P 500, an active fund that masks itself in passivity. The S & P shot gives you immunity from the downside, at least they claim. However, if the index is all you own, it sure cuts you off the upside, as I endlessly prove. The purveyors of conventional wisdom act as if nothing has happened that could make it easier to pick stocks since since they began their insistence on you checking your brain at the door of your savings — nothing like the web, the chatbots, the bountiful information we all know exists but our financial “betters” still ignore. So, out of deference to the creators of “TBOY,” I decided to do more than show up. I listened to old podcasts. And listened some more. And some more — right through the three hours of time I leave for quiet homework, even before Ragu and Toni get up. No, don’t buy Campbell’s because of those hounds. Rest in peace to my old dog, Nvidia. The TBOY podcast was delicious. Just crisp, funny, smart and on point. Just like young people really interested in the markets can give you. Just as young people want the information now, not in ancient and flat form, but in something that’s much harder and more creative with a staccato, machine-gun style of delivery. As I listened to some recent episodes, I heard one that was so spot on that I found myself thinking I should actually highlight some of their analysis on “Squawk on the Street” before I saw them. Oh, by the way, what does “TBOY” stand for? “The Best One Yet.” So, I knew it was right to be going on this show and, more important, I knew there could be no pride of authorship. The boys behind “TBOY” figured out the great conundrum facing this market, which is the existential nature of OpenAI. More specifically, they realized that OpenAI has pledged to spend hundreds of billions of dollars to beat Alphabet -owned Google with ChatGPT. But it can’t. And it won’t. OpenAI, they said, wants to be Google with comprehension, but we don’t need it because we have Google with Gemini. In other words, Google is already everything OpenAI aspires to be. Released on Tuesday, Google’s latest version of Gemini — its AI chatbot to rival ChatGPT — is remarkably capable, with enhanced reasoning capabilities. Additionally, Gemini 3 demonstrates that the scaling laws of AI are still intact, just as Nvidia’s Jensen Huang has for months insisted was the case in the face of some concern about the pace of improvement for AI models. Soon after the Morning Meeting, I went up to see Nick and Jack at their Nasdaq haunt. They were more than gracious and hilarious, frankly, as I thought they would be, as well as respectful beyond all belief, which I found somewhat embarrassing and totally charming. Before we could sit, I complimented them on their triumphant Google observation. As true students of the game of the book tour, though, they preferred to dive into my book. Right from the get-go, minutes after we were mic’d up, they began to press and press about index funds versus picking stocks. They had read the book well, knew it chapter after chapter, as I always hoped would be the case. It was a joy to have actually knowledgeable interlocutors in this, the final station upon my author’s promotional cross. Candidly and somewhat remorsefully, I thought for sure that during my press tour for the book, there would be actually someone who would challenge me, but you can’t challenge me if you haven’t read it. What can I say? It made me rapturous to actually talk about why you can pick stocks, the comparison to when I began to build a portfolio versus now, and how the index fund predators would never let anyone pick a stock, lest they pick the speculative names like Rigetti Computing , Oklo , Joby Aviation and others like it. I, on the other hand, am happy to “allow” readers to own index funds along with self-directed stocks. Why not? Thoughtful investors, armed with the newfound ease of the homework, might select one or two stocks among five that can be life-changing, like Nvidia was to so many of you. The hour flew by. I demanded more time. They thought I was jesting. I was just so damned happy that they got it — it being the revolution I was trying to start when I wrote this book, a rebellion against the index-fund orthodoxy that, at its core, is an insult to the intelligence of everyday people. But no, it was time to depart. I had to write my show and interview a CEO before that. Plus, this was all transpiring on the day the market had a hideous about-face, with none other than Nvidia leading the way into the abyss of an island reversal, up to down in one horrendous session. When I got back, I thought I should write a segment covering what I thought about TBOY and their thesis of OpenAI being beaten by the revitalized of Google. Then I realized, there was not enough time. And it would be way too linear. The fact is, the biggest crisis this market has — the one that may be TWOY — is the hubris of the individual behind ChatGPT, Sam Altman. This supercilious man believes that if he spends enough money that he doesn’t currently have, he can challenge Google in the biggest vertical in the world, information, and that his knowledge factory will top the one in Mountain View, California. We, the users of Gemini 3, now know it will be a tough climb. OpenAI appears so far behind this new Gemini that Altman may have to pivot and go after the verticals of the other hyperscalers: social media or retail and perhaps even enterprise software. There’s only one problem with a potential pivot. No, make that three. First, Meta CEO Mark Zuckerberg has already decided to spend any challenger to death regardless of what it will do to his stock. Social media, with all of those targeted ad dollars, will always be Meta’s turf. Zuckerberg has the firepower to be sure that’s the case. Second, Amazon is always going to win in retail, it’s only real competitor being Walmart . The new initiative toward same-day grocery delivery only widens its moat to defend against challengers. Plus, cloud unit Amazon Web Services, back in growth mode , spins off enough cash to make going against Amazon’s cyber-stores a fool’s errand. Which leaves one other place to go: the enterprise. In the “Oedipus Rex” of our time, Altman may have no choice but to challenge Microsoft at its own game. The 27% stake that Microsoft has in Altman’s entity might not matter to the man who will eventually recognize how cornered he is. Sure, there are other routes for OpenAI. Altman can buy Reddit, a terrific idea if only to block others from that amazing advertising vehicle and its trove of audience-generated content that is great to train models on. The best of Hobson’s choice: Altman could write a check to Apple to make ChatGPT the pre-loaded AI model on its operating systems. The check will have to be a big one as Gemini is the presumed choice. Sadly, at least for the market, I think he will attack every hyperscaler, given his Alex Karp-like ego. Karp is the longtime CEO and co-founder of Palantir . So what happens if Altman does? No single company has that kind of money needed to attack all comers. I think we got a glimpse of what could occur when we got the gaffe of all tech gaffes: OpenAI CFO Sarah Friar uttering the word “backstop” at a Wall Street Journal conference in early November. The quick denouement: Altman spends so much that perhaps a teetering OpenAI becomes a national champion with government-backed loans, the presumption being that President Donald Trump can’t let it fail. A failure this proportion could set back our whole bulwark against the Chinese in an AI race rife with national security concerns. In that situation, everyone makes out well and the market actually soars. I’ll take it. Or, Microsoft, sensing OpenAI’s peril, knows that the true value of OpenAI is now much lower than anyone thinks, so Microsoft crams its child down and buys it for several hundred billion, a totally satisfactory answer even if it means that Nvidia has one less customer. The market is reassured that the spend was all worth it and everything resumes the upward climb. Another possibility: The market stops allowing Oracle to build new data centers and cuts off OpenAI’s credit, with no one coming to its rescue. In that scenario the worry would be awful: wave after wave of companies producing shortfalls as everything is over-built. That is the Thursday scenario, the one that produced that painful Nvidia reversal after its spectacular earnings report the prior evening. I think the repudiation occurred because of a version of what I just traced out. Part of that version included an April 2000 nightmare, that fateful middle of the month tech estrangement when the money poured out of that group and headed to safety stocks like Johnson & Johnson , Coca-Cola and Procter & Gamble , hence our recent buy of the latter because it had been the only one left behind. (Memo to second-guessers: Exiting Johnson & Johnson and Google were huge misses of mine, and I know that well. I just waited for them to come down and they never did). Now we are in a benign period, not that we weren’t when November began and we were told by the calendar investors that we would have a tremendous month. There are plenty of people who still think that we are still in “The Year Of Magical Investing.” These believers will continue to think that’s where we are until the money is taken away, which is what will happen. There are others who are willing to skate past the denouement to where April 2000 resides. There are others who think that they can sell all of the tech giants, except Alphabet and Apple, not a terrible hedge. In the end, though, if things play out as the “TBOY” hosts suggest, we do have to go through some turmoil as OpenAI flails and we wait for the positive – or negative — theses play out. Either way, know this: Alphabet has won in the most logical of battles. Let’s hope that Altman knows Trump and it all works out, as it did with Intel , in the end. (Jim Cramer’s Charitable Trust is long META, AMZN, NVDA, AAPL and PG. See here for a full list of the stocks.) As a subscriber to the CNBC Investing Club with Jim Cramer, you will receive a trade alert before Jim makes a trade. Jim waits 45 minutes after sending a trade alert before buying or selling a stock in his charitable trust’s portfolio. If Jim has talked about a stock on CNBC TV, he waits 72 hours after issuing the trade alert before executing the trade. THE ABOVE INVESTING CLUB INFORMATION IS SUBJECT TO OUR TERMS AND CONDITIONS AND PRIVACY POLICY , TOGETHER WITH OUR DISCLAIMER . NO FIDUCIARY OBLIGATION OR DUTY EXISTS, OR IS CREATED, BY VIRTUE OF YOUR RECEIPT OF ANY INFORMATION PROVIDED IN CONNECTION WITH THE INVESTING CLUB. NO SPECIFIC OUTCOME OR PROFIT IS GUARANTEED.
<
