Google is reportedly preparing to support a multibillion-dollar data center project in Texas leased to Anthropic, with construction loans potentially funding an initial phase exceeding $5 billion. A federal judge has also temporarily blocked the Pentagon from labeling Anthropic a national security risk and cutting off government use of its AI tools, a move the judge described as potentially arbitrary and retaliatory.
Google is preparing to support a major data center project in Texas leased to Anthropic as competition for AI infrastructure intensifies. The project, operated by Nexus Data Centers, could exceed $5 billion in its initial phase.
Google is expected to provide construction loans, according to a report, with a consortium of banks also competing to arrange financing. Anthropic recently signed a lease for the 2,800-acre campus as part of its broader infrastructure tie-up with Google.
Construction is already underway, supported by early-stage debt financing from Eagle Point. The site is expected to deliver around 500 megawatts of capacity by late 2026, with potential expansion to 7.7 gigawatts.
Separately, a US federal judge in San Francisco temporarily blocked the Pentagon from labeling Anthropic a national security risk. Judge Rita Lin granted a preliminary injunction, pausing a directive that sought to cut off federal use of Anthropic’s chatbot, Claude.
The ruling came in a lawsuit where Anthropic argued Defense Secretary Pete Hegseth overstepped his authority. The judge described the government’s actions as “arbitrary” and warned against branding a US company as a threat without a clear legal basis.
The dispute followed a breakdown in negotiations between Anthropic and the Pentagon over military AI use. The company resisted allowing its models to be used for lethal autonomous weapons or mass surveillance.
In her decision, Lin said the government may have retaliated against Anthropic for its public stance. She called the measures a likely violation of First Amendment protections.
US military units reportedly used Anthropic’s Claude AI model during a major airstrike on Iran. This occurred even after a ban order was issued.
