Hacker News new | ask | show | jobs
Can we bootstrap AI Safety despite being unable to even define it? (arxiv.org)
2 points by cryptohell 219 days ago
2 comments
conartist6 212 days ago
AI output is modeled on human behavior. Are humans safe?
link
cryptohell 219 days ago
Given several models, assuming only that some unknown subset is "safe", can we construct a single model as safe as that subset? This reduces obtaining a trustworthy model to a plausibly easier task.
link