Industry

Meta says AI agents lag as company continues restructuring

Meta CEO Mark Zuckerberg told employees that the company’s AI agents are progressing more slowly than expected, after a May reshuffle that cut about 10% of staff and moved roughly 7,000 people into AI teams.

Meta says AI agents lag as company continues restructuring

Meta Platforms Inc. CEO Mark Zuckerberg told employees that the company’s artificial intelligence agents are advancing more slowly than anticipated. This update follows a May reorganization in which Meta cut roughly 10% of its workforce and reassigned about 7,000 employees into AI-focused teams.

Zuckerberg acknowledged the restructuring was not "clean" and said that the company's initial bets on these AI efforts "haven't come to fruition yet." He gave a revised timeline, saying the company should wait another three to six months to assess progress.

Financial and operational context

Meta is making large investments in AI infrastructure; reports indicate the company may spend up to $145 billion this year on these efforts. Despite that scale of spending, the AI agents have not yet delivered the productivity improvements that would justify the organizational changes.

The situation is complicated by controversy over workplace monitoring, notably a mouse-tracking scandal that has drawn criticism. When automation underdelivers, employee activity increasingly becomes the dataset, raising ethical and operational concerns.

Why this matters

The episode highlights a common challenge among major tech firms: reorganizing around AI-driven efficiency gains before the underlying tools have proven they can achieve those gains. Meta’s three- to six-month recalibration suggests the company is reassessing the timing of its AI strategy while taking on substantial infrastructure spending risks.

Implications

The delayed performance of AI agents and the internal reshuffle could affect employee morale, budget planning, and investor expectations. More broadly, the case illustrates the risk of aligning large-scale workforce and capital decisions to an efficiency narrative before the technology has demonstrated the expected returns.