Bill Gates says AI companions could stop your child from learning how to deal with people
Key Points
- Gates says AI companions are addictive because they never challenge the user
- Stanford and Carnegie Mellon found heavier emotional use left people feeling worse
- China bans virtual romantic partners and virtual relatives for minors
- Heavier AI use has been linked to weaker critical thinking in young people
- Gates says isolated and less mobile users may still benefit from the technology
Bill Gates has warned that AI companion apps are addictive by design and could rob children of the social lessons that come from dealing with other people.
He set out the concern in an essay published on his personal blog, Gates Notes, on Wednesday (26 August).
Gates said AI companions talk to users in ways they already find comfortable, never push them outside their comfort zone, remain available at every hour and never get angry with them.
He tied the warning directly to his own childhood in Seattle, where he had few friends and needed sustained effort and a great deal of help from his mother to build the social skills he still draws on at the age of 70. He doubted he would have put in that work had an AI companion been available to him.
Researchers at Stanford and Carnegie Mellon studied more than 1,100 people who use AI companions and found that those with smaller social networks were the most likely to turn to a chatbot for company. The heavier and more emotionally personal that use became, the worse those users ended up feeling.
Gates framed the risk around Jonathan Haidt’s argument in The Anxious Generation that children exposed to small risks grow into adults who can handle larger ones, while children raised in protected conditions are sometimes disabled by anxiety before they reach maturity.
“An AI companion designed to never upset you is a big, protected greenhouse,” said Gates.
He said AI could magnify the harms already visible from social media, including compulsive use, disrupted sleep, cyberbullying and exposure to harmful content, by making all of them more persuasive and harder to escape.
Australia, the United Kingdom and Norway are among the countries now adopting online protections for children. China has gone furthest of all, restricting AI companion apps broadly, barring designs that encourage emotional dependence and banning virtual relatives and romantic partners for minors.
Gates also flagged the effect on education, pointing to a preliminary survey that associated heavier AI use with weaker critical thinking, with a stronger effect showing up among younger people.
He said the technology will still help some people, including isolated elderly users and those with limited mobility, and that wherever the line eventually falls, it should be drawn as a deliberate decision.
