For nearly 10 years, I received a call almost every night between 7:00 PM and 7:01 PM Eastern Time. I didn’t have to look at three letters on my phone screen to know who was ringing. It was more like the old man we called Pop, or “old man of the mountain” as he called himself when he had grandchildren. Sometimes I got tired of the words, but I always took a breath before saying hello so he couldn’t hear the tiredness in my voice from something I knew he would miss one day. “Jamesy, this is the best thing you’ve ever done,” he said. Always “the best thing ever”. If I have any regrets, it’s that I didn’t tape it because I wanted it to play now between 7pm and 7:01pm every night. But I didn’t. So, call me intrigued. I learned from my schedule that I would soon be interviewed by two gentlemen named Jack Krivisi Kramer and Nick Martel on a podcast called “TBOY.” I knew these were the two people who started Robinhood Snacks, which I know. I still read this snack in the morning. It’s around 6:30 a.m. for the group of lazy people I’m dealing with. I had heard about some of their work since then, but frankly, I didn’t pay close attention until I found out I would be interviewed by them on “TBOY.” At least, I wasn’t paying enough attention. I’ve always had an affinity for young people who love the markets, so I thought this interview, this interview, would also be an attempt to convince the general public that they’ve actually read my new book, How to Make Money in Any Market, and can pick a few stocks (five, of course) to go along with the ubiquitous index funds we’re mandated to have, along with the mumps, diphtheria, whooping cough, chickenpox, and measles vaccinations. At a time when so much is up for debate, I have the right to claim that I can buy shares in companies that I can observe. You know, get curious about them, Google them, look at their websites, and discover everything that has given them admission to the hallowed S&P 500, which is often an active fund in the guise of passiveness. At least they claim the S&P shot will be immune to the decline. However, as I have proven endlessly, when the index is all you own, the upside is guaranteed to be cut short. Conventional wisdom purveyors act as if nothing has happened to make stock picking any easier since they started insisting we check our brains at the savings gate. There is nothing like the web, chatbots, and the wealth of information we all know about, but our economic “benefits” still ignore. So, in honor of the creators of TBOY, I decided to do more than show my face. I listened to an old podcast. Then I asked some more. And for the entire three hours I leave to do my homework in peace, even before Raghu and Toni wake up. No, don’t buy Campbell’s because of that hound. I pray that my old dog, Nvidia, may rest in peace. TBOY’s podcast was delicious. It’s refreshing, funny, smart, and to the point. Exactly what a young person interested in the market will give you. Young people now want information not in the old, flat format, but in a harder, more creative, staccato, machine gun format. I listened to a few recent episodes and thought I should actually cover some of their analysis of “Squawk on the Street” before I actually watch it because I heard some that hit the nail on the head. By the way, what does “TBOY” stand for? “The best thing ever.” So I knew it was the right thing to do on this show, and more importantly, I knew there was no way I could have any pride as a writer. The boys behind “TBOY” understood the big challenge facing this market: the essence of OpenAI’s survival. More specifically, they found that OpenAI is committed to spending hundreds of billions of dollars to beat Alphabet’s Google in ChatGPT. But you can’t do that. And it’s not. They said OpenAI wants to be Google with understanding, but we don’t need that because we have Google with Gemini. In other words, Google is already everything OpenAI aspires to be. Google’s latest version, Gemini, an AI chatbot comparable to ChatGPT, released on Tuesday, is much more capable with improved reasoning capabilities. Additionally, Gemini 3 shows that the laws of AI scaling are still intact, as Nvidia’s Jensen Huang has been claiming for months in the face of concerns about the pace of improvement in AI models. Immediately after the morning meeting, I went to meet with Nasdaq’s resident Nick and Jack. Frankly, they were more polite and cheerful than I expected, and incredibly respectful, which I found both a little embarrassing and very charming. Before I sat down, I complimented them on their triumphant Google observation. But as true students of book tours, they preferred to dive into my books. Right from the start, minutes after we got on the mic, they started pressuring us over and over again about index funds and stock picking. They had read the book well and understood it chapter by chapter, as I always hoped they would. It was a pleasure to have a real knowledgeable interlocutor in this work, the final stage of my authorial promotion. Frankly, with some regret, I was pretty sure that some people would actually challenge me during the press tour for this book, but you can’t challenge me if you haven’t read the book. What should I say? I was ecstatic to actually talk about why you can pick stocks, when I started building portfolios versus now, and how index fund raiders don’t let anyone pick stocks, like speculative names like Righetti Computing, Oklo, and Joby Aviation. On the other hand, I am happy to “allow” my readers to own index funds along with voluntary stocks. Why not? Armed with their newfound ease of doing their homework, prudent investors might pick one or two of the five potentially life-changing stocks, as Nvidia was for many. Time passed quickly. I requested more time. They thought I was joking. I was so happy when they figured it out. This was the revolution I was trying to start when I wrote this book, a rebellion against the orthodoxy of index funds, which are essentially an insult to the intelligence of everyday people. But no, it’s time to leave. Before that, I had to write the show and interview the CEO. Moreover, this all happened on a day when the market took a terrifying turn, with none other than Nvidia leading the way into the abyss in one terrifying session. When I get back, I thought I’d write a post summarizing my thoughts on TBOY and its theory that OpenAI was defeated by a resurgent Google. Then I realized I didn’t have enough time. And that would be too linear. In fact, the biggest crisis in this market, perhaps twofold, is the arrogance of Sam Altman, the individual behind ChatGPT. This super smart guy believes that with enough money he doesn’t have right now, he can take on Google in the world’s largest vertical industry, information, and his knowledge factory in Mountain View, California, will be at the top of its game. We Gemini 3 users know it’s a tough climb. OpenAI is so far behind this new Gemini that Altman may need to pivot and pursue other hyperscaler areas: social media, retail, or even enterprise software. There’s just one problem with the potential pivot. No, make it three. First, Meta CEO Mark Zuckerberg has already decided to spend his challenger to death, regardless of the impact it will have on his stock. Social media, with its targeted ad spend, will likely continue to be meta’s turf. Zuckerberg has the firepower to ensure that. Second, Amazon is always going to win in retail, and the only real competition is Walmart. New efforts toward same-day grocery delivery only widen the barrier for challengers. Additionally, cloud arm Amazon Web Services is back in growth mode, spinning off enough money to make it easier to compete with Amazon’s Cyberstore. All that remains is the company. In a modern-day Oedipus Rex, Altman may have no choice but to take on Microsoft with his own game. Microsoft’s 27% stake in Mr. Altman’s business may not matter to Mr. Altman, who will eventually realize how cornered he is. Certainly, there are other routes to OpenAI. Altman could buy Reddit, but it would be a great idea if he just blocked others from its amazing advertising medium and treasure trove of audience-generated content, perfect for training models. Hobson’s best choice: Altman could write a check to Apple to make ChatGPT a preloaded AI model into the operating system. The check should be big, as Gemini will probably be the choice. Sadly, given his Alex Karp-like ego, I think he’ll attack any hyperscaler, at least when it comes to the market. Karp is Palantir’s longtime CEO and co-founder. So what would happen if Altman did that? No single company has the money needed to attack all visitors. I think we got a glimpse of what happens in every technical gaffe. OpenAI CFO Sarah Friar coined the term “backstop” at a Wall Street Journal conference in early November. The easy end: Altman is investing so much money that President Donald Trump probably can’t afford to let it fail, and the struggling OpenAI can become a national champion for government-backed financing. Failure at this rate could undermine the entire bulwark against China in an AI race rife with national security concerns. In such a situation, everyone does well and the market actually skyrockets. I’ll take it. Or Microsoft, sensing OpenAI’s crisis, knows that OpenAI’s true value is now much lower than anyone thinks, so Microsoft packs its baby and buys it for hundreds of billions. That’s a perfectly satisfying answer, even if Nvidia loses one customer. The market is relieved that the spending was worth it and everything resumes its rise. Another possibility: the market will no longer allow Oracle to build new data centers, OpenAI will be discredited, and no one will come to its rescue. In that scenario, the worry would be terrible. Everything is overbuilt, so one company after another creates a deficit. That was the scenario Thursday, and the one that prompted a painful reversal for Nvidia after a stellar earnings report the night before. I think the denial happened because of the version I just tracked. Part of that version included the nightmare of April 2000, the fateful defection of tech players when money flowed out of the group and into safe-haven stocks like Johnson & Johnson, Coca-Cola, and Procter & Gamble, and we recently bought the latter because it was the only one left behind. (Note to second guessers: Leaving Johnson & Johnson and Google was a big mistake for me, and I know it all too well. I was just waiting for them to pull out, and they never did). We’re in a good place right now, but it wasn’t when November started and calendar investors were telling us it was going to be a great month. There are many people who think we are still in the “magical year of investing.” These believers will continue to think that’s where we are until the money is taken away, and that’s what will happen. There are others willing to skate beyond the grand finale to where April 2000 resides. Some believe that selling all the tech giants except Alphabet and Apple wouldn’t be a terrible hedge. But ultimately, if things play out as suggested by the TBOY organizers, we will experience some level of confusion as OpenAI becomes unstable, and we will have to wait for theories to develop, positive or negative. Either way, know this. Alphabet won the most logical battle. Let’s hope Altman knows Trump and that everything works out in the end like it did with Intel. (Jim Cramer’s charitable trusts are long META, AMZN, NVDA, AAPL, PG. See here for a complete list of stocks.) As a subscriber to Jim Cramer’s CNBC Investment Club, you will receive trade alerts before Jim makes a trade. After Jim sends a trade alert, he waits 45 minutes before buying or selling stocks in his charitable trust’s portfolio. If Jim talks about a stock on CNBC TV, he will issue a trade alert and then wait 72 hours before executing the trade. The above investment club information is subject to our Terms of Use and Privacy Policy, along with our disclaimer. No fiduciary duties or obligations exist or arise from your receipt of information provided in connection with the Investment Club. No specific results or benefits are guaranteed.
