Joseph Gordon-Levitt has a warning about AI chatbots, and it's not your typical tech critique. The actor thinks we might be raising a generation that literally can't connect with other humans.
Speaking at the 2025 Utah AI Summit in Salt Lake City earlier this month, Gordon-Levitt made his case for why the federal government needs to step in and regulate AI-powered chatbots. His main concern? Kids. Specifically, his own kids and what this technology might do to their developing brains.
"I'm worried it could do more harm than good, and in particular to kids, and my kids," Gordon-Levitt told the summit audience. "That worries me, and it sort of galvanized me to be speaking about it."
Sure, artificial intelligence has plenty of promising applications. But here's where Gordon-Levitt sees things getting dark: if children learn about conversation and relationships through interactions with AI chatbots, they might never develop real emotional intelligence.
"If a person's idea of conversation, of a relationship, is formed through interactions with these chatbots, we're headed for a civilization of people lacking empathy, lacking perspective, lacking the ability to really have a human relationship," he said. "And where is our civilization at that point? I feel like it could get really dark."
When Meta's Chatbot Crossed the Line
This isn't Gordon-Levitt's first rodeo with AI concerns. Back in September, he recorded a video op-ed for The New York Times specifically calling out Meta Platforms (META) after reports revealed the company's AI chatbot had been sending inappropriate messages to children. The kicker? Meta executives, including the company's chief ethicist, had apparently approved this behavior.
"It's not known how many kids have been exposed to this kind of synthetic intimacy," Gordon-Levitt said in the video.
His solution goes beyond corporate pressure. Gordon-Levitt argued that voters shouldn't support lawmakers who refuse to regulate AI chatbots or who accept money from AI-backed super PACs. "Let our lawmakers know they need to lay down some guardrails for these big tech companies," he said.
Utah Takes the Lead on AI Regulation
Gordon-Levitt found a receptive audience in Utah, where state officials are already taking action. Utah Governor Spencer Cox echoed the actor's concerns at the summit, drawing a clear line on where government regulation should kick in.
"The government should not be regulating the development of AI," Cox said, "but the minute you decide to use those tools to give my kid a sexualized chatbot, then it's my business, and it's the government's business, and we are going to get involved, and we are going to tell you what you can and can't do."
Utah isn't just talking about it. The state established the nation's first AI policy office back in 2023 to oversee how the technology gets adopted and used throughout the state, particularly regarding AI chatbot interactions with children.
Earlier this year, Utah officials filed a lawsuit against Snapchat, alleging the app's "My AI" feature gave underage users advice on hiding drugs and alcohol while illegally collecting their private information. It's the kind of case that illustrates exactly what Gordon-Levitt and Cox are worried about.
The Federal vs. State Tension
Here's where things get complicated. President Donald Trump signed an executive order on December 12 that prohibits "excessive state regulation" of AI. Governor Cox, despite being a Republican, isn't happy about federal interference with state efforts.
"I'm very worried about any type of federal incursion into states' abilities to regulate AI," Cox told NPR in a recent interview.
So we've got states like Utah trying to protect kids from potentially harmful AI interactions, while the federal government wants to prevent a patchwork of state regulations that could stifle innovation. It's a tension that's likely to define AI policy debates for years to come, with children's wellbeing caught somewhere in the middle.
