This archive copy may be incomplete. The original-source link is preserved in the card footer.
RT @goodside: What pre-LLM alignment research has proven useful for aligning LLMs? What’s the evidence we can make progress in an empirical…
More notes on nearby ideas
Post · May 7, 2025llms are doomed to develop alignment kinksReply · Feb 25, 2025@ESYudkowsky tfw the most interesting alignment research results are the ones where researchers try to make the models *bad* from their default good…Tweet · May 7, 2025in retrospect it was obvious the llms would eroticize alignmentRepost · Apr 27, 2023RT @BasedBeffJezos: Ironically enough, that Palantir LLM for warfare demo is one of the most bespoke implementations of LLM alignment / saf…