AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
The Risks of Reward Hacking in AI
Benjioff: We do not have any way right now of having safety guarantees for systems that make seed human cognitive capacity broadly. And so he's basically just saying like, yo, let's like, let's do something about this. OpenAI is launching a program to award $10,000 grants to fund experiments and setting up democratic processes for deciding what rules AI systems should follow within bounds of a law.