Parents could get alerts if children show acute distress while using ChatGPT
(www.theguardian.com)
Seems like a great way to drive people away from using AI, it won't take long for people to abandon a tattle-tail service that they have to worry about self-censoring on.
replyWow, the false positive rate on that will be interesting. from OpenAI's point of view, having it flood the parents with false positives is _way_ safer than having some slip past.
replyso needless to say....
Yeah I had that conversation with Copilot about how to transform into a fox which had some of the best coaching I ever got from a human director but what if it tells somebody I might have STPD?
reply