According to CNBC, investment firm Rothschild & Co Redburn has upgraded Meta’s stock from neutral to a buy rating. Analyst James Cordwell raised his price target to $900 per share from $740, which suggests a potential 37% rally from current levels. He cites Meta’s advertising “demand machine” and expanding AI capabilities as key catalysts for unlocking new revenue. Cordwell specifically mentions AI projects like Andromeda, GEM, and Lattice, and believes the AI reboot aligns with the Nvidia Blackwell cycle. He also points to the acquisition of AI agent firm Manus and the potential in agentic AI and AI video generation. Meta shares have only gained 2% over the past 12 months, making this a bold call for a significant turnaround.
The AI Gap Between Hype and Revenue
Look, the analyst’s logic is clear. Meta has a phenomenal ad business and a ton of AI research. Combining them should print money. But here’s the thing: we’ve been hearing the “AI will supercharge ads” story for a while now. The real question is, when does the lab work translate to the bottom line in a way that justifies a $900 share price? Cordwell admits that current estimates factor in “all the cost and little upside from AI.” That’s a huge red flag. It means the market is already paying for this expensive AI investment, but isn’t yet believing in the payoff. Convincing them will require more than just promising “incremental revenue potential.” It’ll need concrete, quarterly proof that AI is making ads significantly more effective or cheaper to deliver.
The Hardware and Talent Reality Check
The note mentions leveraging hardware advances to improve AI, which is a massive, ongoing cost center. And let’s talk about that Nvidia Blackwell cycle alignment. Sure, new chips are powerful, but so is everyone else’s checkbook. Meta is spending billions on Nvidia AI chips, but so are Google, Microsoft, and Amazon. This isn’t a unique advantage; it’s an arms race where staying in the game is brutally expensive. The comment about AI chief Alexandr Wang being “better qualified than appreciated” feels a bit like trying to spin a narrative, too. Meta’s AI leadership has seen some high-profile departures. Believing in a single executive to close the gap with the frontier by mid-year is a leap of faith, not an investment thesis.
The Agent and Video Pipe Dream
Now, the two “compelling” opportunities—AI agents for SMEs and revolutionizing entertainment with AI video—sound fantastic. In theory. But basically, these are greenfield, unproven markets where Meta has no track record. Creating reliable AI agents for business is a problem the entire industry is struggling with. And AI video? That’s a field crowded with well-funded specialists like OpenAI (Sora) and Runway. Having a strong position in user-generated video on Instagram Reels is not the same as leading in generative AI video technology. These are long-term moonshots, not near-term revenue drivers you bank a 37% stock gain on.
The Bottom Line: Caution
So, is there a case for Meta? Absolutely. Its core ad business is a beast. But this upgrade seems to be pricing in a best-case, frictionless scenario where every AI bet pays off quickly and decisively. History suggests that’s rarely how tech transitions work. The path is littered with delays, technical hurdles, and unexpected competition. For investors, the risk isn’t that Meta fails at AI. It’s that it succeeds, but at a cost and over a timeframe that doesn’t live up to this suddenly rosy $900 expectation. Buying on the promise of FY26 guidance feels like trying to catch a falling knife, hoping the handle is made of pure AI gold.
