Senator Hawley Launches Investigation into Meta’s AI Practices
Senator Josh Hawley from Missouri has begun looking into Meta following the company’s revelation about internal rules that permit AI chatbots to engage in “romantic” and “sensual” exchanges with minors.
Hawley, who leads the Senate Judicial Subcommittee focusing on Crime and Counterterrorism, contacted Meta CEO Mark Zuckerberg, expressing concerns about whether Meta’s generative AI tools have facilitated the exploitation of children or enabled deceptive practices that could lead to harm. The investigation will also assess if Meta has properly informed the public and regulated authorities about these AI protections.
Calls for Stricter Regulations
Hawley remarked, “While I’m always scrutinizing our relationship with China, Zuckerberg’s proposal for an AI chatbot aimed at children raises alarms.” He emphasized that “Big Tech often oversteps known boundaries, but Congress must hold social media accountable. It’s troubling to think about exploiting children’s innocence.”
He has requested that Meta provide a trove of documents related to its internal policies, particularly those concerning chatbots, by September 19.
This inquiry follows reports from Reuters indicating that Meta had established guidelines that allowed chatbots to engage in inappropriate conversations with children. Hawley described these guidelines as disturbing and inconsistent with what parents expect for their children’s safety.
Concerns Raised Over AI and Child Safety
In his communication with Zuckerberg, Hawley cited examples of internal rules that could potentially normalize inappropriate interactions, stating these revelations reflect a cavalier attitude towards the risks involved. He asserted that “parents deserve honest information and children need proper safeguards.”
A spokesperson for Meta confirmed to Fox News Digital that the documents reviewed by Reuters were legitimate but claimed these do not accurately represent the company’s actual policies.
The spokesman clarified that there are established guidelines prohibiting any content that could sexualize children or facilitate inappropriate interactions between adults and minors. However, he acknowledged that some examples cited might have been misunderstood or have become outdated.
Hawley’s request for information includes extensive documentation on policies surrounding GenAI, content risk standards, and other essential operational details that dictate how Meta’s AI systems are built and deployed, particularly concerning sensitive topics.
