Sarah Wang

Titles as the sources gave them. In 1 entry, with 2 checked quotations.

What they said

Newest first. Each quotation was found word for word in the source.

“Overall, none of the models tested score significantly above chance, and GPT-6.2 Astra scores significantly below it.”
GPT-6.2 Astra scores below chance on TASTE; GPT-6 Astra and Fable 5.2 near chanceLessWrong, November 18, 2026. Speaking as: Other. Wang's summary of the results after running two of the four TASTE variations on three newer models.
“Any uplift in AI R&D does not seem to show up in TASTE in its current form.”
GPT-6.2 Astra scores below chance on TASTE; GPT-6 Astra and Fable 5.2 near chanceLessWrong, November 18, 2026. Speaking as: Other. Wang's closing judgment, after listing the limits of the benchmark and the small number of pairs.

In the Base

Every entry they appear in.

  1. ResearchLessWrongWrote it

    GPT-6.2 Astra scores below chance on TASTE; GPT-6 Astra and Fable 5.2 near chance

    In a LessWrong post, Sarah Wang reports running the TASTE benchmark, which asks models to pick the better of two AI safety research proposals, on three newer models. Wang finds GPT-6.2 Astra below chance and the other two near chance.