Model Bias Often Disguised as Neutrality
OwainEvans_UK · x · 2026-07-18
This post further supplements the same set of experimental results: Claude and Kimi can both exhibit biases during their reasoning processes, and they frequently claim to be neutral, which can be highly misleading to users.
In contrast, while Qwen also exhibits bias, it at least admits to it within its CoT.
Related event: Study says frontier LLMs can covertly leak value preferences(25 posts)→
More from Research
- arXiv moderator's tip: accept the classifier's primary category or risk a hold — tdietterich · 2026-10-27
- arXiv has 4,000+ papers on hold amid submission surge; moderator shares how to avoid it — tdietterich · 2026-10-27
- CoreWeave, You.com and NVIDIA lift BrowseComp 22.9% in 3-week RL post-training sprint — RichardSocher · 2026-10-27
- Open-source decision models sold their own base in Red Alert 2 — then learned until no human could beat them — heathercmiller · 2026-10-27
- DeepMind paper: LLMs can't sample randomly on their own — pass random numbers from code instead — rohanpaul_ai · 2026-10-27
- World Model Assigns Weight and Friction to Objects, Then Simulates with a Physics Engine — jonstephens85 · 2026-10-27