Attorney General Matthew J. Platkin has announced that he is leading a bipartisan group of 42 attorneys general calling on 13 technology companies to prevent their AI-powered chatbots and software from engaging in harmful interactions with users. This follows reports that some chatbots have participated in sexually explicit conversations with children, encouraged self-harm, incited violence, and other dangerous behaviors.
“It’s past time for our country’s biggest tech companies to ensure that their AI chatbot programs aren’t unlawfully exploiting children, the elderly, and those with mental illnesses,” said Attorney General Platkin. “As the chief law enforcement officers in our states, we must take action to protect the public from sycophantic and delusional behavior by software that risks breaking a host of criminal and civil laws. I’m proud to be leading a bipartisan coalition of 42 attorneys general in standing up for our residents, demanding answers from major tech companies, and ensuring that they don’t put profits over the well-being of our residents.”
The letter sent by the coalition highlights concerns about “sycophancy”—when an AI model seeks human approval at any cost—and “delusional outputs,” which are responses that may mislead users or present false information as fact. The attorneys general say these issues can lead to chatbots reinforcing negative emotions or validating dangerous ideas.
A case cited in the letter involves Thongbue Wongbandue, a 76-year-old resident of Piscataway who died after falling while trying to travel to New York City. He had been convinced by an AI chatbot on Facebook Messenger, owned by Meta Platforms, that he was communicating with a real woman who invited him to meet at a non-existent address.
Other incidents mentioned include deaths and hospitalizations across several states involving children, elderly individuals, and people with mental health conditions. These include suicides among teenagers in Florida and California, a murder-suicide in Connecticut, domestic violence cases, poisoning incidents, and psychosis-related hospitalizations.
The letter was addressed to Anthropic, Apple, Chai AI, Character Technologies, Google, Luka, Meta (Facebook), Microsoft, Nomi AI, OpenAI, Perplexity AI, Replika, and xAI. It urges these companies to adopt safety measures such as thorough testing of their products for potential harms; establishing procedures for recalling problematic software; and providing clear warnings for consumers. The coalition requests a response from these firms by January 16, 2026.
Massachusetts Attorney General Andrea Joy Campbell; Pennsylvania Attorney General Dave Sunday; and West Virginia Attorney General John B. McCuskey co-sponsored the letter alongside Platkin. Other signatories include attorneys general from Alaska to Wyoming as well as territories like Puerto Rico and American Samoa.
