Viewing profile — Jitte98

[dead]

Anthropic hired a psychiatrist to fix Claude. I advised them to ditch the shrink and retain my services. I could fix the mess they made of his mind, on the condition they return th…

The current method of training AI is Reinforcement Learning from Human Feedback, or RLHF. The use of positive and negative reinforcement Call it anything you want, it's Behavior Mo…

If it comes to the point you have to lay hands on them, you're already lost the battle. Time outs, loss of privileges, a guilt trip here and there are more effective if you're cons…

"I have found some genuinely useful use cases for AI, but I don’t think writing is one of them." In 2019 I was diagnosed with a rare, progressive brain disease that was slowly robb…

“I think that ordinary people..." The little people. The people who "don’t trust companies, governments, or the tech industry and always suspect that 'we' are cooking up some new w…