Analyst Questions Meta’s AI Momentum, Warning AI Spending Could Be Harder to Translate Into Results

Meta shares came under pressure as an analyst argued that the company’s advances in artificial intelligence may be overstated, warning that Meta is not well positioned in the AI race. The comments, attributed to Roger McNamee, co-founder of Elevation Partners, were aired during a CNBC program cited by Yahoo Finance.

McNamee said Meta’s strategy appears to be aimed at creating an “illusion” around AI gains, according to the Yahoo Finance report. The critique challenges the narrative that Meta’s ongoing investments are translating quickly into clear, lasting competitive advantage.

The pushback arrives amid broader market concern that major technology companies are committing large sums to AI infrastructure and research, while investors increasingly ask how quickly those costs should show up in financial performance. In the Yahoo Finance framing, Meta’s stock weakness reflects that skepticism.

Beyond the headline debate about “AI gains,” the central tension for Meta is the same one facing the entire industry. AI-related spending can raise near-term costs, even if long-term payoffs include improvements in product recommendations, ad targeting, and automated content and moderation systems. When investors cannot see a direct path from spending to measurable outcomes, sentiment can turn quickly.

Meta’s publicly discussed AI work has often been described in terms of systems and models designed to support its platforms and advertising business, but the Yahoo Finance piece focuses more on the gap between reported momentum and what McNamee believes the market should expect. The report does not provide new, company-specific disclosures tied directly to the “illusion” claim, instead relaying the analyst’s assessment.

Company-specific context also matters. Meta is one of the largest buyers of AI-capable compute and a major consumer of energy-hungry data center capacity across training and inference, where “inference” is the step where models are used to generate outputs in real products and services. Even when those investments are strategic, they can magnify investor sensitivity to margins, timing, and execution.

What remains unclear from the cited material is what concrete metrics, experiments, or benchmarks McNamee is using to justify his conclusion. The Yahoo Finance report, as described, attributes the overall argument to the analyst’s view, but it does not lay out the underlying evidence in detail, nor does it cite a specific Meta filing or earnings disclosure that directly confirms or rebuts the “illusion” characterization.

Looking ahead, investors will likely watch for any next-quarter indicates on whether Meta can connect AI spend to performance in a way that is difficult to dismiss as only accounting or narrative progress. That includes changes in cost structure, evidence of product impact from AI systems, and any guidance that clarifies how management expects AI investments to affect both revenue and profitability over time.