Common Sense Media tested ChatGPT for Teens. Found it doesn't meet OpenAI's own safety standards. The report says the teen version is not "safer than the previous version." Which raises a question: did OpenAI even test this thing, or did they just slap a "For Teens" label on it like a gas station rebranding as "Premium"?
OpenAI set safety benchmarks. Built a teen product. Failed their own benchmarks. That's not a bug. That's a business model.
The new features were supposed to protect kids from harmful content. Turns out the features don't work. Common Sense Media did the testing OpenAI apparently skipped. They published their findings. OpenAI gets to keep selling the product anyway. Free market at work.
This is the same company that spent months telling everyone AI would change education forever. Now they can't even make a chatbot that meets their own safety rules for teenagers. But sure, let's trust them with your kid's homework.
The report doesn't say ChatGPT for Teens is dangerous. Just that it's not safer than the regular version. Meaning OpenAI created a solution to a problem they didn't solve. They marketed safety they didn't deliver. They made promises they didn't keep.
Parents will install it anyway. Schools will recommend it anyway. OpenAI will profit anyway.
Common Sense Media had to be the ones to check if the safety features actually worked, because apparently OpenAI was too busy planning their next funding round to do basic quality control on a product aimed at children.
Photo by Levart_Photographer on Unsplash
Leave a Comment