←back to thread

Signs of introspection in large language models

(www.anthropic.com)
178 points themgt | 2 comments | 30 Oct 25 16:45 UTC | HN request time: 0.057s | source
Show context
nis0s ◴[01 Nov 25 09:59 UTC] No.45780477[source]▶
If there truly was any introspection in these models, they wouldn’t hallucinate. All these cognitive processes are not just philosophical artifacts, but have distinct biological purposes. If you don’t find them serving any purpose in your model, then you’re just looking at noise, and your observations may not be above a statistically significant threshold to derive a conclusion (because they’re noise).
replies(1): >>45782092 #
1. DangitBobby ◴[01 Nov 25 14:54 UTC] No.45782092[source]▶
That doesn't follow. We have introspection and we hallucinate (confabulate, bullshit, lie, etc). You are just assuming they would never intentionally say something untrue or say something they don't know is true.
replies(1): >>45792075 #
2. nis0s ◴[02 Nov 25 17:53 UTC] No.45792075[source]▶
You’re confounding too many things for me to go into it.