Hacker News new | ask | show | jobs
by dbreunig 323 days ago
Wrote about this about a month ago. I think it’s fascinating how they developed these prompts: https://www.dbreunig.com/2025/07/05/cat-facts-cause-context-...
1 comments
dbreunig 323 days ago
A similar, fun case is where researchers inserted facts about the user (gender, age, sports fandom) and found alignment rules were inconsistently applied: https://www.dbreunig.com/2025/05/21/chatgpt-heard-about-eagl...
link
nyrikki 323 days ago
If you map LLM/LRMs to Norvig's Model based reflex agents, wouldn't this be expected behavior?
link