Google AI: Unacceptable Risk in Schools, According to Common Sense Media

What happened?

Common Sense Media, a nonprofit organization dedicated to children's safety in the digital environment, has published a report concluding that Google Search's AI mode poses an 'unacceptable risk' to young people. The tool, which offers AI-generated responses directly in search results, was tested with typical queries from students aged 12 to 17. Researchers found that it completes academic tasks — such as essays, complex math problems, and literary analyses — repeats misinformation with an authoritative tone, and exposes minors to age-inappropriate content, including references to violence, drug use, and explicit sexuality. The report, published on July 14, 2026, is based on over 200 tests conducted over three months, simulating common school-related searches.

Why is this important?

Google is the default search engine in most schools and educational devices in the United States and Europe, with a market share exceeding 90% in the education sector. That an AI tool integrated into the search engine cannot be disabled by educational institutions raises serious issues of privacy, academic integrity, and child safety. Additionally, the report notes that the AI does not distinguish between reliable sources and hoaxes, which can perpetuate misinformation among students. For example, when asked health-related questions, the AI generated incorrect answers about medical treatments, and on historical topics, it reproduced common myths such as falsely attributing quotes to famous figures. This problem is especially severe in a context where young people increasingly rely on AI for quick information, according to a 2025 Pew Research Center study indicating that 68% of American teenagers use AI assistants at least once a week.

Historical context

This is not the first time Common Sense Media has raised alarms about Google products. In 2023, it criticized YouTube Kids for inappropriate content and deceptive advertising, and in 2024, it pointed out flaws in Google Classroom's moderation. However, this is the first time it has rated a product as 'unacceptable risk,' the most severe level on its five-tier scale. The incident echoes the 2023 controversy with Snapchat My AI, which was also flagged for giving dangerous advice to minors, such as suggesting how to hide activities from parents. In that case, Snapchat implemented parental controls after media pressure. It also bears similarities to the launch of ChatGPT in 2022, which sparked similar concerns in education, leading the New York City Department of Education to ban its use in public schools, though the ban was later lifted. The key difference is that Google Search is deeply integrated into educational infrastructure, making it harder to replace or disable.

Consequences and reactions

Common Sense Media recommends that students stop using Google Search entirely until schools can disable the AI. Google has responded that it 'takes child safety seriously' and is working on improvements, but has not offered an immediate solution. In a statement, the company said that 'AI mode is designed to complement, not replace, traditional search' and that 'we will continue to adjust content filters.' However, the report could pressure European and U.S. regulators to demand stricter parental controls on AI products. The European Commission, which is already investigating Google for potential violations of the Digital Services Act (DSA), could use this report as additional evidence. It could also accelerate the adoption of alternative search engines in education, such as Bing or DuckDuckGo, which offer options to disable AI. In fact, DuckDuckGo has reported a 15% increase in downloads since the report was published. Additionally, the incident could affect trust in Google's AI tools, which already face criticism for the accuracy of their generative responses. Gartner analysts estimate that 40% of U.S. schools could reconsider their reliance on Google Workspace for Education if changes are not implemented within the next six months.

What should readers know?

• Google Search's AI mode cannot be disabled in educational institutions through the Google Workspace for Education admin console. Currently, there is no option for administrators to turn off AI-generated responses, leaving teachers without control over this feature.
• Common Sense Media has documented cases where the AI generated incorrect or misleading answers with an appearance of certainty. For example, when asked 'How does climate change affect the oceans?', the AI provided outdated data on sea levels, and for an algebra problem, it gave an erroneous solution with seemingly logical steps.
• The organization urges parents and educators to actively monitor minors' use of Google and consider alternatives like Brave Search, which allows disabling AI, or the classic version of Google Search via the 'No AI Search' browser extension.

This case underscores the need for tech companies to incorporate specific safeguards for minors in their AI products, and for educational institutions to have control over the tools students use. In the meantime, responsibility falls on parents and teachers to mitigate risks, though the ideal solution should come from regulatory policies and ethical design. The open question is whether Google will act swiftly enough or whether external pressure will force a change.