Hacker News new | ask | show | jobs
by skywhopper 16 days ago
Your “some methodological deficits” is doing a lot of work.
1 comments

What if the methodological deficits are actually causing the paper to underestimate the quality of the AI responses? Why assume any deficits would bias the AI's competence upwards instead of downwards?
Why not assume the AI is god and the rapture is happening tomorrow?