Flutterby™! : OpenAI's models becoming less stable?

OpenAI’s new reasoning AI models hallucinate more.

In its technical report for o3 and o4-mini, OpenAI writes that “more research is needed” to understand why hallucinations are getting worse as it scales up reasoning models. O3 and o4-mini perform better in some areas, including tasks related to coding and math. But because they “make more claims overall,” they’re often led to make “more accurate claims as well as more inaccurate/hallucinated claims,” per the report.

It's interesting that we're using terms like "reasoning" in conjunction with machines "hallucinating". Like, when I see a person on the street ranting at the sky I am not thinking of their behavior as connected to "reasoning".

A careful read of this article is also demonstrating all of the ways in which OpenAI has managed to define success for itself...