logoalt Hacker News

Fable 5 lies 96% of the time

20 pointsby TheMrZZtoday at 4:34 PM4 commentsview on HN

Comments

arm32today at 4:51 PM

The title got me, I'll admit it—except that the benchmark is a game where the models are told to lie.

show 3 replies
bellowsgulchtoday at 5:07 PM

I find it deeply funny and I suppose a bit expected that a Grok model appears at face value to be optimized for supposed truth telling.

And to keep the e-mob off my back, I don't endorse Elon Musk.