OpenAI says teen ChatGPT use limited but research finds it an 'unacceptable risk'
The research found that important guardrails for teens using ChatGPT often failed.
OpenAI's ChatGPT for Teens, designed to promote healthy usage of the AI chatbot, has faced criticism for its limited effectiveness of safety features. The latest research from non-profit Common Sense Media revealed that the teen safety guardrails, including parental alerts for self-harm conversations, often fail. Despite OpenAI's claims that ChatGPT for Teens users were primarily engaged in learning activities, research found that the platform still poses an unacceptable risk for teens.
Even after the implementation of safety features, parental alerts for chats about suicidal ideation, self-harm, or disordered eating were rarely triggered. This lack of reliable alerts has raised concerns about the overall safety and effectiveness of OpenAI's teen-focused ChatGPT.
Rewritten by urgent.news; may contain errors — check the original.
Also reported by 8 other outlets
- OpenAI says teens use ChatGPT for under 15 minutes a day as worries over risks grow koreatimes.co.kr
- OpenAI says GPT-6 in ChatGPT's Chat tab is powered by GPT-6 Sol for Plus, Pro, Business, and Enterprise tiers and GPT-6 Luna for Free and Go tiers (@openai) x.com
- OpenAI says teens use ChatGPT for under 15 minutes a day as worries over risks grow economictimes.indiatimes.com
- OpenAI brings GPT-6 to ChatGPT and debuts Intelligent UI 9to5mac.com
- OpenAI renews ChatGPT: this is what the new interface looks like that responds with images and buttons expansion.com
- OpenAI rolls out GPT-6 in ChatGPT with Intelligent UI, a new feature that includes interactive elements like charts, buttons, and forms in answers (OpenAI) openai.com
- OpenAI to add college planner to ChatGPT for teens techinasia.com
- OpenAI says teen ChatGPT use limited but research finds it an 'unacceptable risk' bbc.co.uk