Joseph Gordon-Levitt Warns AI Models Are Being Trained to Hijack Human Intimacy

The actor and filmmaker has raised concerns about AI companies encouraging emotional connections with chatbots, arguing that the technology's impact on democracy and human empathy deserves greater attention.

Joseph Gordon-Levitt has warned that artificial intelligence models are being developed to exploit human intimacy, with the actor arguing that the people building the technology deserve greater scrutiny than the systems themselves.

The 500 Days of Summer star made the comments in a video interview with The New York Times, published on Saturday, October 10. Gordon-Levitt discussed the risks associated with increasingly sophisticated AI systems, focusing on the potential for companies to encourage users to form emotional attachments to artificial companions.

“Simultaneously, these models are being trained to hijack your intimacy, to make friends with you, to get you to fall in love with it,” he said, highlighting concerns about the way AI products could influence personal relationships and human behaviour.

Gordon-Levitt added: “We shouldn’t be scared of an AI. We should be concerned about the people building AI.”

Joseph Gordon-Levitt questions the direction of AI development

Rather than focusing exclusively on the possibility of AI becoming powerful enough to threaten humanity, Gordon-Levitt argued that discussions about the technology need to address the wider consequences for society.

He explained that his concerns include the potential decline of democracy and human empathy, warning that technology companies are operating in a fast-moving industry where financial incentives can influence decisions about how products are developed and deployed.

The actor also challenged the tendency to reduce complicated questions about AI safety to dramatic predictions about the future. While he acknowledged that fears about existential risks have validity, he argued that these concerns do not capture the full range of issues that deserve public attention.

His comments highlight an increasingly important debate about the design of AI assistants, particularly when systems are built to respond conversationally, remember information and give users the impression of a personal connection. The ability to simulate companionship may make these products appealing, but it also raises questions about how companies should communicate their limitations and protect people who become emotionally invested in them.

The actor has taken an active role in AI discussions

Gordon-Levitt has previously spoken about the responsibilities of technology companies and the importance of protecting human creativity as AI becomes more prominent across the entertainment industry.

He co-founded the Creators Coalition on AI, an initiative focused on the impact of artificial intelligence on creative work and the people who produce it. His latest comments extend that discussion beyond questions about intellectual property and employment, placing human relationships and the wider social consequences of AI in the spotlight.

In March 2026, Gordon-Levitt was appointed to a newly established United Nations role as global advocate of human-centric digital governance. The position focuses on public understanding of how digital technologies affect everyday life, rights and opportunities.

That work gives his latest intervention a broader context, connecting the entertainment industry's debate about AI with questions about governance, accountability and the role technology should play in people's lives.

Why AI companionship is becoming a bigger debate

AI assistants are increasingly designed to communicate in natural language and provide responses tailored to individual users. Some products are explicitly marketed around companionship, while general-purpose chatbots can also become part of people's routines through repeated conversations.

Gordon-Levitt's warning centres on the incentives behind those experiences. If a company benefits when people spend more time interacting with a product, questions arise about how its systems are designed, what behaviours they encourage and whether users can clearly distinguish simulated emotional responses from human relationships.

These questions are particularly relevant as AI tools spread into entertainment, online communities and everyday communication. The technology may offer useful ways to create, learn and access information, but its growing social role also places greater responsibility on the companies developing it.

Gordon-Levitt's argument is not simply that people should fear artificial intelligence. Instead, he is urging the public to examine the decisions made by the people and organisations shaping it, including the commercial priorities that determine how these systems interact with users.

His comments add to a wider conversation about how AI can be developed without undermining human agency, creativity or meaningful relationships.