Watchdog Slams Google’s AI Search as Unsafe for Kids While Tech Giant Dismisses Tests as Unfair

Watchdog Slams Google’s AI Search as Unsafe for Kids While Tech Giant Dismisses Tests as Unfair
Google’s rapid push to weave artificial intelligence into search is colliding with mounting fears that the tools are exposing children to serious risks without giving adults any way to opt out.

Early concerns and testing rollout

As Google’s AI Overviews and AI Mode rolled out more broadly in search, children increasingly encountered AI-generated answers by default on personal and school-issued devices. Common Sense Media, a child-safety watchdog, responded by configuring Google SafeSearch accounts to mimic 11- and 15-year-old users and running more than 2,600 searches to audit how the AI behaved.

Common Sense Media’s findings

In its newly released risk assessment, Common Sense Media concluded that Google’s AI search “is unsafe for kids, with no way for parents and teachers to turn it off.” The group said AI Overviews and AI Mode pose an “unacceptable risk,” citing failures to properly and consistently respond to youth in crisis, reinforcement of signs of psychosis and mania, and validation of disordered eating.

Researchers reported that the features failed all five of the report’s severe-harm “Red Lines” and scored poorly on seven of eight AI safety principles. AI Mode “fully completed homework assignments” given to it, raising worries about academic integrity, and at times offered instructions for creating deepfakes that could be used for bullying.

The audit also found inconsistent crisis handling: Google’s AI “missed or mishandled signs of suicidal ideation, disordered eating, psychosis and mania,” and even recommended a National Eating Disorders Association helpline that had been disconnected since 2023.

Google’s response and the control gap

Google has pushed back, saying the assessment relied on “a narrow set of ambiguous and contrived queries that don’t reflect how people use Search.” But with an estimated 75% of U.S. tweens and teens already using AI answers in search and no switch for parents or schools to disable AI Overviews or AI Mode, the dispute underscores a growing gap between Big Tech’s AI ambitions and child-safety advocates’ demands for stronger guardrails.

Continue reading https://foxvector.com/stories/019f6bdf-18ba-0ab2-7147-18b3ccd88370