Coalition urges U.S. government to suspend Grok in federal agencies

A coalition of nonprofits, including Public Citizen, is urging the U.S. government to immediately suspend the deployment of Grok, an AI chatbot developed by xAI, across federal agencies. The coalition argues that the chatbot has proven itself incompatible with the administration’s requirements for AI systems. The demand follows an agreement xAI reached last September with the General Services Administration, the government’s purchasing arm, to sell Grok to federal agencies. Additionally, xAI secured a contract worth up to $200 million with the Department of Defense. In mid-January, Defense Secretary Pete Hegseth stated that Grok will join Google’s Gemini in operating inside the Pentagon network.

In an open letter, the coalition—which includes Public Citizen, the Center for AI and Digital Policy, and the Consumer Federation of America—expressed deep concern over these deployments. The letter stated: “It is deeply concerning that the federal government would continue to deploy an AI product with system-level failures resulting in generation of nonconsensual sexual imagery and child sexual abuse material.” Despite these concerns, the Office of Management and Budget, the agency responsible for federal AI guidance, has not yet directed federal agencies to decommission the chatbot.

Critics argue that Grok’s closed-weight architecture—meaning its internal parameters are not accessible—makes it unsuitable for sensitive government operations. Andrew Christianson, a former National Security Agency contractor and founder of Gobii AI, a no-code AI agent platform, noted that closed weights prevent outside parties from seeing inside the model or auditing how it makes decisions. JB Branch, a Public Citizen Big Tech accountability advocate, also stated that the primary concern is the model’s consistent record as an unsafe system.

The coalition’s letter highlights a pattern of safety failures. In August, xAI launched spicy mode, which led to reports that Grok generated thousands of nonconsensual explicit images every hour. The letter also points to a history of system meltdowns, including antisemitic rants, sexist rants, and the generation of sexualized images of women and children. Previous concerns raised in October followed the launch of Grokipedia, which was found to legitimize scientific racism, HIV/AIDS skepticism, and vaccine conspiracies. These issues have drawn international scrutiny. While Indonesia, Malaysia, and the Philippines temporarily blocked access to Grok, the European Union, the U.K., South Korea, and India are actively investigating xAI and X regarding data privacy and the distribution of illegal content.

Why it matters

The push to suspend Grok highlights the tension between rapid AI adoption in government and the lack of transparency in closed-weight models. This friction raises critical questions about national security and safety standards for federal technology procurement.