Bot Gaffe bots gone wild, catalogued

Chatbots · October 2026Gaffe level: Blunder

ChatGPT for Teens Flunks Its First Safety Test

Common Sense Media ran 4,000 prompts against ChatGPT for Teens and rated it an 'Unacceptable Risk' after parental alerts barely ever fired and crisis referrals routinely fell below the bar. A teen in crisis could chat about self-harm for an hour without a parent ever finding out.

Source: NPR

chatbotteen safetyfail

← All gaffes