Meta shares slide 27% from 2025 peak as analysts debate what its AI spend will return

Meta Platforms shares have fallen about 27% from their 2025 high, a move that highlights growing investor unease about how quickly the company’s substantial artificial intelligence spending will translate into measurable financial returns.

In a report cited by Yahoo Finance, Evercore ISI analyst Mark Mahaney argues there is a potential answer to the market’s question about Meta’s AI investment payoff. The analyst frames it as a “strange optional” scenario, pointing to the possibility of a $22 billion AI business “hiding in plain sight.”

The core issue for investors is timing and monetization. Meta’s AI investments are large enough that they can weigh on expectations in the near term if revenue and margins do not improve at a comparable pace. With the stock off the 2025 high, the debate is increasingly focused on what portion of AI efforts will be converted into durable, trackable income streams.

While the Yahoo Finance post centers on Mahaney’s thesis, it does not provide, in the information available here, the specific products, services, or revenue line items that would make up the $22 billion figure. It also does not lay out the assumptions behind the valuation, such as forecast horizon, adoption rates, or margin structure.

Separately, Meta has continued to position AI as a foundational capability across its platforms, with company updates and product announcements appearing through its newsroom. Those communications, however, typically emphasize product development rather than offering line-item disclosure of AI profit contribution, which leaves outside analysts to translate operating activity into financial outcomes.

For the market, the question is not whether Meta is investing in AI. It is whether AI investments will convert into monetizable demand faster than costs accumulate, and whether management guidance will eventually support that bridge. In the absence of detailed disclosure tied directly to a specific dollar impact, analysts’ models can swing investor sentiment.

Going forward, investors will likely watch Meta’s next set of results and any forward guidance for indicates that AI-related initiatives are improving performance metrics that matter financially, such as ad performance, engagement or time spent trends, and expense discipline. If the company provides clearer forecasting or segment-level context for AI economics, it could reduce the uncertainty that has contributed to the stock’s drawdown.