Before AI Meets Kids, It May Need a Crash Test
Synopsis
AI safety for children is facing greater scrutiny as usage expands, with former Google executive Tom Siegel seeking independent testing and UNICEF reporting widespread adoption across 10 countries.
AI safety for children is under a new spotlight after the call from former Google vice president trust and safety, Tom Siegel, for independent safety testing of AI products used by children.
In May, Common Sense Media launched its Youth AI Safety Institute to study children's use of AI, and set safety standards, with Siegel to be the first leader of the new organization.
The former Google executive said companies should enhance age verification and introduce external evaluations for AI systems. Siegel proposed that companies could test AI products like cars: they undergo safety testing before being allowed to drive on public roads.
86% of U.S. Children Have Used AI
Common Sense Media conducted its most recent survey, Common Sense Media Census: AI Use by Tweens and Teens (2026), with 1,204 U.S. children ages 9-17.
The study found 86% have experience using AI, and nearly a quarter use it daily. More than 40% said they have never talked with a parent or caregiver about safety concerns related to AI.
One out of every six children who encountered an AI chatbot saw content they considered to be inappropriate for their age, according to the same research. Over half of AI users were seeking information about their health or bodies.
The findings serve as a backdrop to Siegel's worries about the safety of AI for children, including risks of suicide, psychosis, and children becoming overly dependent on AI, affecting their critical thinking.
UNICEF Finds 20 Million Children Using AI
UNICEF found that at least 20 million children had used AI, based on a statistical analysis of nationally representative household surveys in 10 countries.
The countries were Armenia, Brazil, Colombia, the Dominican Republic, Jordan, Mexico, Montenegro, North Macedonia, Pakistan and Serbia. The data do not indicate a ranking for a country, but rather the level of adoption across multiple countries.
According to UNICEF, more than 2 million children have used AI to ask for advice on matters of interest to them, representing one in every 10 kids, and around 13 million children have consulted AI to learn and to do homework.
Children were also using AI more than three times as often as adults in the countries covered, the agency added.
In a statement, Google has highlighted that parents can block their child's access to Search, and OpenAI has said that it uses age-reporting, account and behavioural data to ensure that a child is not underage. Anthropic did not immediately respond to a request for comment.
The new institute will develop its own criteria for measuring and evaluating the safety of publicly available AI products, with AI safety for children at the center of their testing process.
Source: Reuters
Pooja Malik is a business journalist with over six years of experience covering startups, entrepreneurship, and emerging trends. She has previously worked with leading media platforms such as YourStory Media and BW BusinessWorld, where she reported on business, policy, and market developments. Currently, she serves as Editor at The Inspirepreneur Magazine, where she writes and edits stories across business, lifestyle, and travel, with a focus on clarity, accuracy, and reader relevance.
You Might Also Like
India Inflation Drops 0.25%, RBI Relief Ahead
Meloni and Trump Meeting Signals New Phase in US-EU Tariffs Debate
