Meta Platforms is reportedly negotiating a deal to rent AI data center capacity to Anthropic worth up to $10 billion over two years. The potential agreement highlights Meta's strategy to monetize its extensive AI infrastructure investments by offering excess compute capacity to other firms. Anthropic pitched the arrangement in June, and both sides could exit early, with terms remaining fluid, according to a report by The New York Times on Friday citing sources familiar with the discussions.

The proposed deal is smaller in scale compared to Anthropic's recent agreements. It is roughly a third the size of Anthropic's $45 billion, three-year computing deal with Space Exploration Technologies Corp. signed in May. CEO Mark Zuckerberg stated in May that Meta plans up to $145 billion in 2026 capital spending, more than double 2025's $72 billion. He indicated that the company could sell excess capacity if it overbuilds, noting that Meta has not yet engaged in cloud services because it prioritizes its own compute needs.

Meta has previously established similar compute partnerships. In April, the company struck deals with CoreWeave, and in March, it secured an arrangement with Nebius Group N.V. These transactions align with Meta's broader approach to leveraging its data center assets. Meanwhile, Anthropic's web traffic share nearly doubled between March and June, separating it from smaller rivals and strengthening its negotiating position as it seeks more compute capacity to meet demand.

Trading Metrics

Meta's market capitalization stands at approximately $1.64 trillion. The stock has traded within a 52-week range of $520.26 to $796.25. Year-to-date, the tech stock has fallen 0.68%.

A U.S. judge has rejected a bid by 26 employees of Meta Platforms to block the company from laying them off while they pursue claims that AI-powered tools targeted them due to disabilities or medical leave. U.S. District Judge William Orrick in Oakland, California, ruled on Friday that an emergency order was not justified because the workers could not demonstrate that losing their jobs constituted the irreparable harm required to halt the process. The layoffs are scheduled to begin on July 22.

The legal challenge centers on allegations that Meta relied on algorithmic performance tools, including AI token-usage dashboards and productivity metrics, to select employees for termination. The plaintiffs argue that these systems disproportionately affected workers who missed work due to medical conditions or family care, effectively penalizing them for exercising protected leave rights. The lawsuit claims this is the first case against a major U.S. company to challenge the alleged use of AI in conducting layoffs.

Meta has denied the allegations, maintaining that workforce decisions were made by people rather than algorithms. The company notified nearly 8,000 employees, or approximately 10% of its global workforce, in May regarding the job cuts. The reduction in staff is part of a broader restructuring effort as the company doubles down on investments in artificial intelligence.

Judge Orrick determined that the merits of the workers' novel legal claims must be decided in private arbitration rather than through the court injunction sought by the plaintiffs. The case highlights growing scrutiny over the role of automated systems in critical human resources decisions.

The dispute underscores tensions between corporate efficiency drives and employment protections. Meta has stated that organizational decisions are human-driven, rejecting claims that AI tools were the primary determinant in the layoff selections.