Senate Staff Get Three Approved AI Tools But Agentic Systems Stay Banned

Senate lawmakers and their staff are barred from all but the most basic AI tools, even as Congress debates how to regulate the technology. An internal communication obtained by NPR, sent to staff in the spring by the sergeant at arms, who oversees chamber cybersecurity, lists three AI chat interfaces free of charge to Senate offices.

Three tools allowed under strict limits

The approved services are Microsoft Copilot Chat, Gemini Chat for Google Workspace Enterprise Plus and OpenAI ChatGPT Enterprise. Under the rules, the platforms "cannot independently access internal Senate drives, shared folders, email, Teams chats, or other Senate resources."

The sergeant at arms has not authorized the more advanced tools at the center of the debate in Washington, among them OpenAI's Codex and Anthropic's Claude Code and Cowork. Such systems let AI-powered agents complete tasks autonomously at a user's request. The office describes the approach as an effort to balance innovation with risk.

Vetting continues, with no timeline

Advanced AI tools are being vetted for specific use cases, a Senate staffer told NPR on condition of anonymity; no timeline was offered for that work.

"Congress, unfortunately, has always lagged behind," said Adam Kovacevich, founder and CEO of the tech policy group Chamber of Progress. He added that lawmakers are "writing rules for technology they, in many cases, never even used and that's a problem."

Daniel Schuman, executive director of the nonpartisan American Governance Institute, said tools able to exfiltrate data pose a significant risk for the Senate and the House. Still, he said, "those concerns are not the best reason for the non-approval of certain technologies."

The House and the committee's answer

Across the Capitol, the House runs a separate system overseen by the Office of the Chief Administrative Officer. A House staffer said its rules outline specific approved use cases, with tools from Microsoft, OpenAI, Google and Anthropic permitted in various scenarios. The office declined to comment.

The top Republican and Democrat on the Senate Committee on Rules and Administration did not address the policy's details. A spokesperson for committee Democrats said the rules are meant to "protect Senate data and include stronger security and contractual requirements."

Some offices already use AI informally. Sen. Bernie Moreno, Republican of Ohio, told the technology show MTS Live that his office has an AI staffer it named Albert, used for research, though Senate rules limit wider deployment. Kovacevich warned: "if your only mental model of the technology is the harm, you're probably going to write a bill that's all brakes and no steering wheel."