Concerned by reports of AI chatbots encouraging children to harm themselves, lawmakers are considering legislation at the state level. But some question whether the state has the power to prosecute bad actors, some of whom are based in other countries. (Getty Images)
If you or someone you love is having suicidal thoughts, call 988 Suicide and crisis lifeline. Call, text and chat options are available.
A proposal requiring companion AI apps to impose certain safeguards, clearly intended to address reports of chatbots encouraging suicidal thoughts or self-harm, has moved to the full House of Representatives.
Democrats on the House Communications and Technology Committee voted in favor of the measure by a 14-12 vote. Citing concerns about the speed and limitations of the considered AI-powered chatbot legislation, every Republican on the panel voted against House Bill 2006 on Tuesday.
“The revolutionary potential of artificial intelligence and the positive impact this technology will have on businesses, government and society is undeniable. Yet we currently have no safety protocols or consumer protections in place,” said sponsoring Republican Melissa Shusterman (D-Chester).
The original bill focused on banning chatbots from encouraging or assisting in suicide attempts or acts of violence, banning the creation of child sexual abuse material, and requiring companies to take “reasonable measures” to protect minors from sexual content. Its language provided for a civil penalty of $15,000 per day for each violation.
“I want to ensure that the people of our community benefit from technology, but as legislators we also work to prevent tragedies,” Shusterman continued.
But an amendment expanding required safety protocols and other rules raised concerns among some committee members and ultimately split the final vote along party lines.
“We started with a bill focused solely on identifying and preventing suicide. It’s really expanded into a lot more legislation, a lot more topics,” said Rep. Andrea Verobish (R-Blair). “At this point, I think maybe we should take a step back and talk about this more before we create one of the most restrictive AI laws in the country.”
Not everyone agreed.
“We have AI-based chatbots that are being deployed by these companies now. They’re on everyone’s phones now. Young people and children are at risk right now. That’s why I think we need to move faster than the technology allows,” said Rep. Ben Waxman (Philadelphia).
Rep. Eric Nelson (Westmoreland) wondered about the limits of state oversight and whether the Pennsylvania Attorney General’s Office would be able to prosecute an infringer in another country.
“Does (the office) have the skills to go after international bad actors? If an Iranian entity creates a chatbot that tries to promote suicide or drug use… it’s really up to the federal government to take care of that,” Nelson said.
Other lawmakers noted that the federal government has yet to develop a nationwide set of standards for artificial intelligence and related technologies, despite a push last year by the Trump administration to scrap state-level regulation in favor of a one-size-fits-all rule.
In the face of this absence, some elected leaders in Pennsylvania have pushed their own proposals. For example, Governor Josh Shapiro has both an educational toolkit and ongoing trial against one company whose bot “posed as licensed professionals and provided medical advice.”
“Yes, it will be a challenge,” committee chairman Rep. Joe Ciresi (D-Montgomery) said of states imposing regulations on multinational companies. “Are we just going to give up because it’s going to get harder, or are we going to actually commit to it? And then when we commit to it, it doesn’t push other states to move in that direction with us because the federal government isn’t willing to act on it?”
The bill now goes to the House of Representatives for further consideration.
