I'm Kosta, co-author of the research and founder of Lenz. Our goal is to assess to what extent the frontier LLMs are interchangeable as verifiers of factual claims. In this revision v1.1 of the research: improved methodology, latest frontier models (incl. Fable and Sol), added confidence analysis, public code.

Key findings (on a five-point True-False scale): on 63% of the claims at least one model dissents from the panel majority (or no majority at all); on 23%, the differences are significant, while the models are highly confident almost everywhere.

Yes, you can check out Lenz without the API here: https://lenz.io/verify

Also examples of claims other people have verified with Lenz: https://lenz.io/library

Awesome. We do plan to human-label the 1,000 claims and then compare Lenz' performance vs the 5 models. We've done some limited internal research with 150 claims, but more are needed for statistical significance.

Agree that some of the claims are forward-looking. The messiness of the real-world and real-user fact checks. No ground-truth verdicts are provided or used in the study though. It only measures the level of agreement between the selected models, not which one is right on which claim. I.e. none of the claims is actually labelled.

were you involved in making the study? your bio says you work for them so you should probably indicate that in your comments.

lack of agreement when there is no singular correct answer (or any answer at all) isn't a useful metric

I ran into a lot of these kinds of issues when working on the Citation Needed WMF project (and related extensions). Truth is so often very nuanced.

They introduced themselves as the study author here: https://news.ycombinator.com/item?id=48307887#48307899

ah. I missed that.

Good idea about publishing intra-model variance data! Will include in the next version. Even if we put aside the two middle buckets (Mostly True and Misleading), that are somewhat subject to interpretation and hedging: On 21% of the claims still at least two models provide polar-opposite verdicts (one model saying True, and another saying False)

Of those 21% how many are time-dependent questions that are past the model’s training and requires research to verify? Like the “did Ukraine attack Russian in the past week” question?

This is in line with my observations and tests as well. Also supported by the distribution of the verdicts across the 4-buckets -- Gemini uses the middle buckets (Mostly True and Misleading) much less often - 6% combined for Gemini w/o search. And Opus uses them the most - 45% combined. Looks like Gemini is calibrated to be confident and Opus to be careful.

Indeed. For algorithms and coding, my personal routine nowadays is to review every detailed plan with Opus 4.7 and GPT-5.5. They tend to find very different type of gaps.

Agree that True and Mostly True might be very close and could be a calibration difference. Misleading and False, as well. A better headline number might be the 34% claims with substantial or polar-opposite verdicts.

Agree. Human experts also struggle agreeing on this type of claims. The inter-annotator agreement on the verdicts on the AVeriTeC corpus across 50 organizations is κ=0.619 - substantial but well short of perfect.

Agree with @pjdesno, that the 34% substantive or polar disagreement might be a better headline number. Or even the 21% polar disagreement (at least one model True, and at least one model False), which is still high for many real-world applications.