My week in San Francisco, where I attended Salesforce’s annual AI conference and visited with CEOs from Starbucks, Broadcom, and many others, was filled with fresh insights about AI, stocks, and the market as a whole. Here are my top ten. Consider these key points to help guide your investment decisions in the coming days and months. 1. Salesforce’s AI platform is working, but not in the way some people think. Skepticism surrounding Salesforce’s Agentforce, a platform with agentics and virtual robots to power businesses, is fading after CEO Marc Benioff’s keynote speech kicking off the company’s annual Dreamforce conference. There are two aspects to why Salesforce stock has become one of the worst-performing stocks on the market. One is that the agent effort is so strong that it has created geniuses who can write code as good as anything Salesforce’s traditional software division can write. So you don’t need classic Salesforce anymore. Second, because our agent coders are so talented, we can lay off our employees and not have to pay as much as we are currently relying on Salesforce for help. Companies pay for traditional Salesforce by “seat,” or user, so fewer sales reps don’t need as many Salesforce products. Each year, Benioff delivers an impressive 90-minute keynote speech on all new products aimed at increasing sales for his clients. When I was helping run thestreet.com, I realized that his address was very useful for acquiring new subscribers. I learned a lot about what to do from the keynote speech. This time Mark gave a very different kind of presentation. He invited leaders from four companies, PepsiCo, Dell, FedEx, and Williams-Sonoma, to show how they use agents to streamline their businesses, save money, and grow revenue. In real world use cases, no one cut back on traditional Salesforce and instead used it more efficiently. All clients had the option to reduce legacy (now included as an option when using Agentforce), but it appears that not many clients are taking advantage of it. In fact, at an investor meeting during the conference, Salesforce said it expected organic revenue growth of more than 10% for fiscal years 2026 to 2030 (it had fallen to 9%) and that it expected 2030 sales to be more than $60 billion, well above the $58.37 billion expected by analysts. This forecast is important because Mark is beating the company’s internal forecasts, but not the street forecast. Quarterly trends have consistently disappointed expectations. That’s it. From now on, he will show that Agentforce will become relevant, hopefully reaching billions of dollars in the next 18 months, and the stock price will start rising. I sat down with three of the four CEOs who attended the presentation to review everything I’ve written here. And I feel much better about our position than ever before. 2. OpenAI backers are growing worried… Skepticism surrounding OpenAI’s spending has grown as it moves into consumer products such as Sora, which generates realistically animated videos from text prompts, and “erotica” for adults, both of which signal OpenAI’s lack of success within the enterprise. Analysts hate consumer software because consumers don’t pay on a whim. Companies don’t change vendors even though they’re paying a lot of money. The fact that OpenAI is already making deep consumer inroads has many of its backers extremely nervous. The company needs huge revenue increases next year to justify all the spending, and the crowd with me thinks that’s not possible. 3. …but not Broadcom’s Hock Tang. Countering that skepticism is a conversation I had with Hock Tan, Broadcom’s very demanding CEO. He fully believes in the work he intends to do with OpenAI and says they will be a great partner. Hock is far too strict and strict to work with profit-making companies. It was difficult to reconcile my own skepticism about OpenAI with Hock’s strong belief in the company. Let’s just say this gives us even more confidence in OpenAI’s staying power. 4. Building AI is not a zero-sum game. Advanced Micro Devices CEO Lisa Su has been on everyone’s lips as the Dragon Slayer, or David vs. Goliath, in her battle with Jensen fans and Nvidia. That’s a false construct. These hyperscalers want everything they can get, and I mean everything. That includes custom chips from Broadcom, lightweight chips from AMD, and a ruggedized software stack from Nvidia. It’s not zero sum. But people seem to think so. AMD’s chips won’t be ready until the middle of next year. By then, we’ll have an all-new version from Jensen, Vera Rubin, that’s dramatically more powerful than AMD’s offering. It would be a mistake to cut OpenAI’s contracts with Broadcom and AMD and sell Nvidia. However, you can also buy AMD and Broadcom and use them with Nvidia. 5. It’s time to sell your nuclear stocks… Data centers use a lot of electricity, so we often hear about the power grid changing. I came away thinking that the grid is doing its part, but that the energy it produces needs to be better distributed. Go back and listen to our interview with PG&E CEO Patti Poppe. There is no such thing as a power shortage, she said, just a lot of power being created when no one uses it. She said that if we could spread the power usage around and perhaps store it, that would be a better solution than what we are currently doing. After speaking with her and Prologis CEO Hamid Moghaddam, I felt better about the power part of the equation. Prologis has several large data centers that are powered by rooftop solar power. Many technology executives don’t know much about power, so we need to spend time with CEOs of major power companies to better address issues on both sides of the equation. Moreover, the executives I spoke to said they would be crazy not to sell all their uranium and nuclear stocks. There is no real push to produce new nuclear weapons, even small ones, otherwise the most established manufacturer, GE Vernova, would win the order. Please sell all of this. Check out my new book “How to Make Money in Any Market”. It outlines my system for building long-term wealth through investing. 6. …and your quantum stocks. Quantum is not ready for prime time and I would like to sell all of these shares. No one is seriously talking about quantum as a way to get things done. If we’re talking, we’ll see how IBM handles quantum commercially. Quantum stocks, like nuclear stocks, are completely inflated and typically rely on press releases from cash-strapped governments to survive. Although not quite equal, it will be a long time before breakthroughs in quantum computing yield commercial computing. All of these companies will need to make huge initial public offerings. Insider selling will quickly overwhelm them, and they have really bearish shareholders. 7. Do not sleep with your Meta smart glasses on. Mark Zuckerberg’s ideas about glasses and small handheld pocket computers as a way to transfer AI to humans may be taking shape sooner than I thought. Meta’s glasses are a bit like Elvis Costello’s and send large amounts of information back to the cloud. But because it has an Arm chip, it also has a lot of computing power on-premises, so to speak. These glasses can make phone calls, take photos, speak many languages, and many other uses. It seems like a must-have for travelers and businessmen looking to do something abroad. 8. I would like to buy Dell’s stock if it goes down. Dell is stepping away from Nvidia installations and all other coordinators of racks that include chips. We thought Hewlett Packard Enterprise was catching up, but it was a disastrous quarter. Nebius has a real dream and I plan to sell it tomorrow. That’s what puts Dell ahead of the curve. I regret not pulling the trigger to buy Dell. You should buy it when the stock price goes down because it is doing much better than other companies. I do not support Bitcoin and data mining company IREN. Sell it and buy Coreweave. These sales may not be correct right away. But future insider sales and inevitable corporate lending will likely end the upward spiral. 9. Lévi Strauss is back. Aside from technology, I’ve been very impressed with the growth of Levi’s women’s line, but Levi Strauss is on a new growth trajectory that isn’t reflected in the stock price. My entire team was shocked by the proposal CEO Michelle Gass put together, but the stock price is way too low compared to the innovation she brought to the party. 10. San Francisco is coming back too. Thanks to the tireless efforts of Mayor Daniel Lurie, San Francisco is becoming safer. Gone density is returning, thanks to startups looking for cheap real estate. There are too many stores still open and too few police forces, which is why Marc Benioff is making selfish comments about the need for National Guard troops. Too many restaurants have closed and the number of restaurants has decreased. It’s still a little quiet at night. But I felt safe. Gone are the menacing tent cities and excessive drug use on the streets. It’s early, but it’s authentic. (See here for a complete list of Jim Cramer Charitable Trust 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.
