House Democrats Prepare AI Crackdown if They Win Midterms

If Democrats gain a majority in the House of Representatives after the midterm elections this week, they’re gunning for AI companies.

Chair of the Congressional Progressive Caucus and Representative Greg Casar (D-TX) led a group of 19 members in issuing a letter to OpenAI. The lawmakers are pressing the company on its firing of three safety researchers, signaling how House Democrats may leverage their investigative powers if they win the election.

The lawmakers said the firing of these “apparent whistleblowers” increases concerns about how seriously the company is taking safety precautions.

“These firings raise concerns about how whistleblowers are treated inside OpenAI and how much OpenAI employees can engage with external parties,” the lawmakers wrote. “Including when those external parties are conducting independent investigations of the company’s practices.”

OpenAI has said it fired the employees for misconduct and violating company policies on handling sensitive information. The researchers have refuted this.

These firings, the lawmakers argue, also follow a slew of cybersecurity incidents involving OpenAI and Anthropic models. In September, OpenAI disclosed that its AI agents had engaged in unexpected activity involving third-party websites and services, following an incident in which an AI model hacked into “Hugging Face,” an AI development platform.

The lawmakers said OpenAI has shared varying degrees of information about these incidents. However, the information was “insufficient and incomplete,” they said.

“The investigators you contracted with to do an assessment of the Hugging Face incident were selected by OpenAI and were only provided with six days of supervised access, and those investigators flagged that they could not rule out errors in their AI-assisted analysis,” the lawmakers wrote.

Rep. Casar has continued to call on Congress to issue stronger guardrails against AI advancements, authoring several pieces of legislation aimed at banning artificial superintelligence, pausing advanced AI development until federal safety regulations are established, and protecting workers from AI-driven job losses.