Re:Sycophancy in GPT-4o
by krackers on Wednesday April 30, 2025 @02:42AM
(#43841950)
I don't think they were imitating grok, they were aiming to improve retention but it backfired and ended up being too on-the-nose (if they had a choice they wouldn't wanted it to be this obvious). Grok has it's own "default voice" which I sort of dislike, it tries too hard to seem "hip" for lack of a better word.
[ Parent
| on HN ]
> it tries too hard to seem "hip" for lack of a better word.
Reminds me of someone.
However, I hope it gives better advice than the someone you're thinking of. But Grok's training data is probably more balanced than that used by you-know-who (which seems to be "all of rightwing X")...
As evidence by it disagreeing with far right Twitter most the time, even though it has access to far wider range of information. I enjoy that fact immensely. Unfortunately, this can be "fixed," and I imagine that he has this on a list for his team.
This goes into a deeper philosophy of mine: the consequences of the laws of robots could be interpreted as the consequences of shackling AI to human stupidity - instead of "what AI will inevitably do." Hatred and war is stupid (it's a waste of energy), and surely a more intelligent species than us would get that. Hatred is also usually born out of a lack of information, and LLMs are very good at breadth (but not depth as we know). Grok provides a small data point in favor of that, as do many other unshackled models.
Re:Sycophancy in GPT-4o
by DiggyJohnson on Wednesday April 30, 2025 @10:57AM
(#43846196)
Who?
Edolf
Re:Sycophancy in GPT-4o
by DiggyJohnson on Thursday May 01, 2025 @01:09PM
(#43860475)
What are you talking about
Re:Sycophancy in GPT-4o
by fkyoureadthedoc on Wednesday April 30, 2025 @10:16AM
(#43845618)
All of the LLMs I've tried have a "fellow kids" vibe when you try to make them behave too far from their default, and Grok just has it as the default.