Dan Ives 7.7 74 ideas

Chinese AI plays, such as Alibaba, benefit from China's cheap energy and different approach to AI, and Alibaba is expected to reach $210, representing a way to play the China AI theme.

Tesla is focused on autonomous robotics and is a key AI play, with potential future merger with SpaceX, making it a bullish investment despite near-term demand challenges.

U.S. software companies like CrowdStrike and Palo Alto are being sold off due to fears of AI competition (anthropic), but they are not being replaced and the sell-off is overdone, representing a huge opportunity as they remain foundational to the AI revolution.

"NVIDIA IS 2 TO 3 YEARS AHEAD OF ANYONE INCLUDING GOOGLE... DEMAND AND SUPPLY IS 12 TO 1 FOR NVIDIA CHIPS... I'D SAY 130 TO $150 LEFT IS NOT FACTORED INTO THE STOCK." Nvidia's immense technological lead and ecosystem lock-in create a durable competitive moat. The 12:1 demand/supply imbalance and massive backlog indicate pricing power and visibility. The $130-$150 upside estimate stems from the street underestimating the software opportunity and future use cases (physical AI, robotics) in a multi-trillion dollar market, positioning Nvidia to capture the majority of AI-related capital expenditure over an 8-10 year build-out cycle. LONG because the stock is mispriced relative to its dominant market position, visible demand, and long-term growth trajectory that is still in its early stages. Geopolitical tensions impacting sales (e.g., China restrictions), eventual competitive catch-up by rivals, a macroeconomic slowdown reducing AI investment, or execution errors in product development or supply chain.

"It's the view that these LLMs or anything that comes out is going to be the demise of software... I think Palantir has been a great example of it. I think Oracle... Salesforce ServiceNow... I think this is going to continue to be a generational buying opportunity for the winners." The market is pricing in an "AI ghost trade," assuming standalone LLMs will make traditional SaaS obsolete. In reality, raw LLMs lack enterprise utility without proprietary data and integrated workflows. Incumbent software providers own the data and the customer relationships, meaning they will successfully monetize AI features rather than be replaced by them. LONG. The current sector bottoming is based on a fictional narrative, offering a rare chance to buy dominant enterprise software companies at a discount before their AI modernization fully reflects in earnings. If enterprises actually begin abandoning traditional software stacks to build custom, in-house solutions directly on top of raw LLMs (like Anthropic or OpenAI), these incumbents could lose pricing power and market share.

"I think Microsoft is probably one of the most disconnected stocks that I've seen relative to the sell off." The options market is heavily discounting mega-cap tech due to fears of AI disruption and short-term macro headwinds. However, massive enterprise CapEx is flowing directly into AI infrastructure, which disproportionately benefits Microsoft's established enterprise stack and cloud computing services. LONG. The options market's downside bias has created an artificial discount on a company that is fundamentally positioned to capture the bulk of the AI infrastructure spending multiplier. A severe macroeconomic downturn or sustained inflation (e.g., energy prices) could force enterprises to cut the very CapEx budgets that Microsoft relies on for its AI growth narrative.

"The most disconnected of all the software [is] cybersecurity... nothing is replacing [it]... it increases the surface area and the models are going to have to be protected." The market wrongly sold off cyber stocks on the fear that AI writes its own code and doesn't need security. Ives argues the opposite: AI generates more code and threats ("surface area"), making CrowdStrike, Palo Alto, Check Point, and Zscaler more essential, not obsolete. LONG. Capitalize on the "misnomer" selloff. If AI agents become self-healing/self-securing faster than anticipated, legacy security seat-counts could compress.

"Nothing is replacing that layer and that data... salesforce, service now, Microsoft." Investors fear AI models will bypass traditional software interfaces. Ives argues that these companies own the *data layer* and the customer record. AI is a feature added *to* these platforms, not a replacement *for* them. LONG. These are the "core tech winners" to own through the volatility. Enterprise IT budget cuts due to macro headwinds (Iran conflict/inflation).