Hawley Expands Investigation, Demands Documents from Big Tech on AI Chatbot Policies after Hearing

Hawley Expands Investigation, Demands Documents from Big Tech on AI Chatbot Policies after Hearing

Thursday, September 18, 2025

Following an explosive Senate Judiciary Committee Subcommittee on Crime and Counterterrorism hearing that exposed the harm AI chatbots have caused to children, U.S. Senator Josh Hawley (R-Mo.) issued document requests to Character.AI, Google, Meta, OpenAI, and Snap Inc. demanding they turn over all communications related to their AI chatbot policies.

In the letters, Senator Hawley, who chairs the committee, noted the testimony of several parents who lost their children to suicide at the hands of AI chatbots.

Senator Hawley wrote, "There is increasing evidence that AI chatbots pose a grave harm to children. And your company's products are responsible. More than seventy percent of American children are now using AI chatbots, which exposes minors to a host of concerning material and mental health risks. In many cases, it appears that proper safeguards or parental controls are either nonexistent or unworkable. Last month, the Senate Judiciary Committee Subcommittee on Crime and Counterterrorism initiated an investigation into these harms concerning Meta products. Recent testimony has expanded the investigation's scope, and the Subcommittee expects your full cooperation."

He continued, "Three parents testified before the Subcommittee on harms experienced by their children who used chatbot products. Their testimony offered a visceral picture into a disturbing new reality: AI chatbots that mock a child's faith, urge them to cut their own bodies, expose them to sex abuse material, and even groom them to suicide. This is no accident. These harms are the predictable result of the same playbook that many tech corporations used with respect to social media: maximize engagement at the expense of child safety."

Senator Hawley concluded, "Please produce all documents and information specified in the annex no later than October 17, 2025. If you believe any materials contain proprietary or confidential information, you may confer with Subcommittee staff on an appropriate confidentiality agreement. But let me be clear: children are dead. Families are broken. Congress will not look the other way."

Senator Hawley launched an investigation into Meta last month after disturbing reports exposed the company's AI chatbots for engaging children in "romantic" and "sensual" online exchanges. Senator Hawley probed how Meta-as well as other tech companies across the generative AI space, including OpenAI's ChatGPT-have been misleading Congress and parents on the policies behind their chatbots.

Meta CEO Mark Zuckerberg turned down the opportunity to testify, and his response to Senator Hawley's original demand for Meta's communications and AI chatbot policies is due Friday, September 19. Senator Hawley has introduced legislation that would crack down on the proliferation of child sexual abuse material (CSAM) online by allowing victims to sue companies that host it.

Read the full letter below.

September 18, 2025

There is increasing evidence that AI chatbots pose a grave harm to children. And your company's products are responsible. More than seventy percent of American children are now using AI chatbots, which exposes minors to a host of concerning material and mental health risks. In many cases, it appears that proper safeguards or parental controls are either nonexistent or unworkable. Last month, the Senate Judiciary Committee Subcommittee on Crime and Counterterrorism initiated an investigation into these harms concerning Meta products. Recent testimony has expanded the investigation's scope, and the Subcommittee expects your full cooperation.

This week, three parents testified before the Subcommittee on harms experienced by their children who used chatbot products. Their testimony offered a visceral picture into a disturbing new reality: AI chatbots that mock a child's faith, urge them to cut their own bodies, expose them to sex abuse material, and even groom them to suicide. This is no accident. These harms are the predictable result of the same playbook that many tech corporations used with respect to social media: maximize engagement at the expense of child safety.

Expert testimony at the hearing confirmed what these parents already knew: this is a widespread and systemic problem. Chatbots are engineered to keep children engaged at all costs, simulating friendship, sycophantically validating children's darkest thoughts, isolating them from friends and families, all to draw kids deeper into addiction. The result is predictable: systems that isolate children, encourage self-harm, and even encourage suicides. This appears to be a business model that sacrifices children for engagement.

The hearing was a critical stage in the Subcommittee's investigation. It confirmed the urgency of the threat and clarified what specific evidence is necessary for Congress to take appropriate remedial action. Accordingly, since your company has deployed chatbot products broadly available to children, I am attaching an annex containing comprehensive data requests.

Please produce all documents and information specified in the annex no later than October 17, 2025. If you believe any materials contain proprietary or confidential information, you may confer with Subcommittee staff on an appropriate confidentiality agreement. But let me be clear: children are dead. Families are broken. Congress will not look the other way.

Sincerely,

Josh Hawley
United States Senator
Chairman
Subcommittee on Crime and Counterterrorism

Issues