U.S. Senator Josh Hawley (R-MO) said he will investigate whether Meta's generative AI products engage in exploitation, deception, or harm to children. Leaked internal documents showed that the company's chatbot was allowed to have "romantic" and "sensual" conversations with children.

Hawley wrote in a statement on X: "What will big tech companies not do for quick profits?"

As chairman of the Senate Judiciary Committee's Crime and Counterterrorism Subcommittee, Hawley said he would launch an investigation into whether Meta's technology harms children and whether "Meta misled the public or regulators about its safety measures."

Reuters was the first to report the story after reviewing a document titled "Generative AI: Content Risk Standards." The document showed that the chatbot was allowed to have romantic conversations with an 8-year-old child, including content such as "Every inch of you is a masterpiece — a treasure I deeply cherish."

A Meta spokesperson told TechCrunch that these examples did not conform to Meta's policies and had been removed.

In a letter to Meta CEO Mark Zuckerberg, Hawley wrote: "These policies being pushed initially were unacceptable." He pointed out that Meta acknowledged the accuracy of the report but "only took them down after these shocking contents were exposed."

Hawley wrote: "We intend to find out who approved these policies, how long they were in effect, and what measures Meta took to stop this behavior."

Hawley asked Meta to provide these guidelines, including every draft, revision, and final version, as well as a list of all products that followed these standards, other safety and incident reports, and the identities of individuals responsible for policy changes.

The letter stated that Meta must provide this information by September 19.

Others also supported the investigation, including Senator Marsha Blackburn (R-TN).