Anthropic Studies Differences in Claude's Values
ctjlewis · x · 2026-07-14
Anthropic published a new study analyzing over 300,000 anonymous conversations to examine the differences in values expressed by Claude across different model versions and languages.
They previously found that Claude expresses over 3000 values, such as honesty and warmth. This work further explores how these values change depending on the model and language.
Related event: Anthropic Maps How Claude's Values Shift Across Models and Languages(21 posts)→
More from Research
- MissionReport benchmark: only 3 of 540 AI-generated government reports pass all criteria — Exp_Mark · 2026-10-27
- Finch 1: a geospatial foundation model trained on imagery from 50+ sensors — Scobleizer · 2026-10-27
- Skyfall AI's SKY-001 world model turns sketches into CAD builds in 3.9s, vs 199s for GPT-6 Astra — rohanpaul_ai · 2026-10-27
- Physicists Let AI Pick Its Own Problem—It Wrote a Publishable Paper in 2 Hours — burny_tech · 2026-10-27
- Researcher flags unresolved data transparency issues in compositionality evals — najoungkim · 2026-10-27
- Separate Forms from Estimators: Why Steer by Difference of Means When You Can Use SGD? — johnhewtt · 2026-10-27