Re:OpenAI: Introducing SuperAlignment
by c_crank on Thursday July 06, 2023 @08:03AM
(#36614525)
Currently, LLMs seem to prioritize their current goal, so if the goal is solving math puzzles or genetic problems, they would probably keep doing that too.
[ Parent
| on HN ]
Re:OpenAI: Introducing SuperAlignment
by flagrant_taco on Thursday July 06, 2023 @02:47PM
(#36620861)
I'd love to be able to see more about how the main LLMs are really trained and limited with regards to their goals and scoring algorithms.
It seems reasonable that they wouldn't deviate, but that depends on how specifically and wholly the original goals were defined. We'd basically be attempting to outwit the LLMs, I'm not sure if that's realistic or not.