3 ms·
They basically defined AI safety as "AI shouldn't say bad words or tell people how to do drugs" instead of actually making sure that a sufficiently intelligent
by ric2b 3y ago
They basically defined AI safety as "AI shouldn't say bad words or tell people how to do drugs" instead of actually making sure that a sufficiently intelligent AI doesn't go rogue against humanity's interests.
- SpicyLemonZest 3y agoI'm not sure where you're getting that definition from. They have a team working on exactly the problem you're describing. (https://openai.com/blog/introducing-superalignment https://openai.com/blog/introducing-superalignment)
- timeon 3y ago> getting that definition from That was not about actual definition fro OpenAi but about definition implied by user Legend2440 here https://news.ycombinator.com/item?id=38344867 https://news.ycombinator.com/item?id=38344867
- ric2b 3y agoSure, they might, but what you see in practice in GPT and being discussed in interviews by Sam is mostly the "AI shouldn't say uncomfortable things" version of AI "safety".