Meta's AI Spending Rattles Investors and Global Markets

Here is a number worth sitting with: a 10% single-day drop in Meta's share price, according to The Mercury News, represents one of the sharpest single-session retreats for a company that touches roughly three billion people's daily lives. Not three billion shareholders. Three billion users — the majority of them outside the United States, many of them in markets where Meta's platforms are not just social media but the practical infrastructure of small business, journalism, and political communication.

When Wall Street reprices Meta, it is not just adjusting a valuation multiple. It is sending a signal about capital allocation priorities that will eventually reshape the product experience for a market trader in Lagos who runs her wholesale business through WhatsApp, or a garment exporter in Dhaka who acquires customers exclusively through Facebook ads. That is the story underneath the earnings drama. The shareholder revolt is the mechanism; the consequences are much wider.

What Actually Happened

BBC News and Yahoo Finance both report the same trigger: Meta shares plunged on Wednesday as investors balked at the company's promise to keep spending heavily on artificial intelligence projects even as profits dwindle. CEO Mark Zuckerberg has positioned the AI push as existential — not optional — and has floated the idea of selling AI tools to other businesses as a future revenue stream. Investors, judging by the reaction, were not persuaded that this constitutes a near-term answer to the cost question.

Bloomberg Tech framed this as part of a broader reckoning across big tech, with AI capital expenditure under scrutiny at both Meta and Microsoft heading into earnings season. The divergence in how markets have received these bets at different companies is worth attention — as covered in rival AI bets from this week, the market is not uniformly punishing AI spending; it is punishing AI spending that it cannot yet map to a revenue model.

That distinction matters. Meta's core business — digital advertising — is already a finely tuned machine. The question investors are asking is not whether AI is a good idea in the abstract; it is whether spending at this scale, with this degree of uncertainty about the payback timeline, is the right use of capital when the advertising business itself faces headwinds from slowing user growth and increasing competitive pressure from TikTok and YouTube.

The Metaverse Echo

There is an obvious historical parallel here, and it is worth being careful with it. When Meta pivoted hard into the metaverse — rebranding the entire company around a bet that had not yet shown commercial legs — investors eventually penalized it severely. The company's stock subsequently mounted a steady recovery toward its all-time high, according to Quartz, though that recovery reflected a confluence of factors: aggressive cost-cutting, a rebound in the digital advertising market, and the market's repricing of Meta's AI narrative itself. It was not simply a reward for retreating from the metaverse.

The pattern worth extracting from that episode is not that Zuckerberg always eventually gets vindicated. It is that Meta's recoveries have consistently been driven by the advertising engine reasserting itself — not by the moonshot bets delivering early. If AI spending is compressing margins on the core advertising business without yet generating its own revenue stream, investors are essentially being asked to finance an R&D operation inside a company they bought as an advertising platform.

The Rate Environment Is Not Helping

The macro backdrop makes this harder. FT News Briefing reports the Federal Reserve is holding rates steady as inflation fears persist. That is not neutral terrain for a company asking investors to accept near-term earnings compression in exchange for long-run AI payoffs. Higher-for-longer rates raise the discount rate on future cash flows — which is a technical way of saying that promises about revenue five years from now are worth less today than they would have been in a low-rate environment. Zuckerberg's pitch requires patience from a market that the rate environment is actively rewarding for impatience.

This is the context in which Zuckerberg's announcement that Meta plans to sell AI tools to other companies landed so flat. It is not an implausible strategy — enterprise AI services are a real market — but it is a long-cycle revenue bet, and the Fed's posture means the market's tolerance for long cycles has shortened.

The Geography Wall Street Is Ignoring

Here is where I want to pull the lens back from the earnings call narrative, because the Wall Street framing captures only part of what is actually at stake.

Meta's advertising revenue is global. In Southeast Asia, sub-Saharan Africa, and South Asia, Meta's platforms — Facebook, Instagram, WhatsApp — are not peripheral products used by a portion of the population. For many small and medium businesses in those markets, they are the primary customer acquisition channel, the customer service interface, and in some cases the payment infrastructure. When Meta faces pressure to cut costs or reallocate capital toward AI infrastructure and away from product development in lower-revenue-per-user markets, the consequences fall first and hardest on the advertisers and users in those regions — who have essentially no voice in the shareholder conversation.

There is also a regulatory dimension with direct financial implications. The European Union's constraints on cross-border data flows and its General Data Protection Regulation enforcement have already limited Meta's ability to use European user data to train AI models at scale. That is not a minor footnote. Europe remains one of Meta's highest-revenue-per-user geographies, and if the AI training pipeline is materially restricted there, the return-on-investment calculus for the AI buildout becomes harder to close — precisely the problem investors are already pricing.

The irony is pointed: Meta's AI ambitions may be structurally constrained in the markets where it earns the most, while the markets where it has the most users — and arguably the most social influence — are the ones where AI-driven product changes will be felt most acutely by people with the least leverage to push back.

A Regulatory Footnote That Is Not Actually a Footnote

Peripherally, CNBC reports the FTC has sued telehealth company Hims and Hers over alleged health data sharing with Meta and Snap, among other billing and subscription practices. The lawsuit sent Hims shares down 10%. This is a separate story in the immediate sense, but it is a data point in the longer arc: the regulatory pressure on how Meta handles third-party data is not confined to Europe. American regulators are beginning to trace data flows through Meta's advertising infrastructure, and that has pricing implications for the ad product itself.

If advertisers — particularly in sensitive categories like health and finance — face increasing friction around targeting and data use, Meta's core revenue model absorbs the cost. That is a different kind of AI investment risk than the one investors are focused on, and it may be the more immediate one.

What the Next Earnings Report Will Actually Tell Us

Zuckerberg's argument is structurally coherent: AI infrastructure is a necessary investment to stay competitive, AI tools sold to enterprises are a new revenue stream, and the advertising business will be improved by AI-driven optimization of ad targeting and content ranking. None of that is implausible. The question is sequencing and cost.

What analysts will be watching in the next earnings report, per Bloomberg Tech, is whether the capital expenditure trajectory shows any inflection — any signal that the spending curve is bending toward a payoff phase rather than continuing to steepen. A credible path to monetizing the AI infrastructure, even an early one, would change the conversation considerably.

Until that evidence arrives, Wednesday's share move is best understood as a quarterly referendum on a promise — not a judgment of record.

---

By Raj Mehta, Global Markets & International Finance Reporter