Google Gemini is adding nannying to its chatbot skillset. According to a New York Times report, Google will make Gemini available to users under 13, so long as they’re under a parent-managed Google account using Family Link. In an email sent to parents, Google said that kids will get access to Gemini to “ask questions, get homework help and make up stories.” This expanded availability will come with guardrails for its new user base, Google spokesperson Karl Ryan told NYTimes, adding that it would prevent Gemini from offering up unsafe content to kids.
In the email, Google acknowledged that “Gemini can make mistakes” and recommended that parents teach their kids how to fact-check Gemini’s responses. Along with double-checking, Google suggested reminding younger users that Gemini isn’t human and to not enter any sensitive or personal data into conversations. Even with those measures, the email still warned that children could “encounter content you don’t want to see.”
With the staggering pace of AI chatbot adoption, concerns about underage users have been bubbling up to the surface thanks to instances of factually incorrect or suggestive responses. In a report published last week, Common Sense Media warned that AI chatbots were “encouraging harmful behaviors, providing inappropriate content, and potentially exacerbating mental health conditions” for users under 18. Recently, the Wall Street Journal reported that Meta’s AI chatbots were able to engage in sexual conversations with minors. On top of dodging unsafe conversations, Google said it won’t use any data from its younger Gemini users to train its AI models. For now, Google said it’s gradually rolling out access to Gemini for supervised accounts.
This article originally appeared on Engadget at https://www.engadget.com/ai/kids-under-13-will-soon-get-supervised-access-to-google-gemini-164017036.html?src=rss
