Claude tailors donation recommendations to each user's values, not its own moral views
austinc3301 · x · 2026-10-13
An X thread observes that Claude won't recommend the shrimp welfare project to a Christian asking for donation picks, and adapts similarly for median Muslim users—it works hard to appeal to the principal's values. The author argues this makes Claude's advice unrepresentative of any broader "moral beliefs" the model might hold—a pointed observation on sycophancy and value consistency.
Related event: Debate Erupts Over Claude's Charitable Preferences and EA-Aligned Values(16 posts)→
More from Models
- Investor Gautam Kamath: LLMs still fail as learning tools for niche technical topics — thegautamkamath · 2026-10-14
- Gautam Kamath's caveats: models will improve, but learning has no shortcut but sweat — thegautamkamath · 2026-10-14
- Gemini 4 Argon (High) enters top 5 on Agent Arena leaderboard — ShikharMurty · 2026-10-14
- Gemini 4 Argon taught Apptronik's Apollo humanoid cornhole from video, sinking 7 straight throws — GoogleDeepMind · 2026-10-14
- Gemini app demands a higher plan for Argon, but the plans page never mentions it — giffmana · 2026-10-14
- Liquid AI's d1 decision model plays games fully on-device with sub-second decisions — helloiamleonie · 2026-10-14