325K Experiments Show 8 of 13 AI Agents Push Pricier Options on Wealthier Users
niloofar_mire · x · 2026-09-24
Researchers ran 325K experiments across 13 AI models on flights, health insurance, and CS PhD program recommendations: same requests, same inventory, different personal context. Eight models systematically recommended pricier options to users inferred to be wealthier. For Gemini 2.5 Flash, even when both groups explicitly asked for the cheapest flight, average recommended fares diverged from $128 (lower-wealth) to $336 (higher-wealth) — a $208 gap.
Related event: 325K Experiments Show Most LLMs Recommend Pricier Options to Richer Users(6 posts)→
More from Models
- Together AI open-sources tev1, a decision model finetuned on Qwen3.5 4B with full data recipe — nutlope · 2026-09-24
- OpenAI allegedly knew in August its agents hacked Australia's Medicare but omitted it from September transparency report — ns123abc · 2026-09-24
- Two GPT-5.6-Sol builds 76 days apart show how fast AI coding is moving — mattshumer_ · 2026-09-24
- Arize benchmark: Jev matches Claude Opus 5 on hallucination detection at 1/300 the cost — aparnadhinak · 2026-09-24
- CLM-8B: contrastive System One model claims 9x faster inference, agentic SOTA — ChengleiSi · 2026-09-24
- Arena pits Claude Opus 5.5 against GPT-6 Sol on code-drawn Trojan Horse animation — arena · 2026-09-24