Harvard-MIT paper: "plan ahead" prompts make agents play worse; interface design matters more
dair_ai · x · 2026-10-06
A Harvard-MIT study tests whether adding "plan ahead" or "think about the other players" prompts actually helps agents — in auctions and matching markets with known optimal strategies, so every choice can be scored.
Key findings:
- Those planning prompts make play worse overall
- Changing the interface worked better: an ascending auction presenting one safe choice at a time reduced bid errors across four model families, and plainly describing payoffs and why truthful bidding is safe also helped
- Agents' stated plans didn't track their choices: interventions that improved bids left measured reasoning quality unchanged, and some prompts improved plans without improving bids
The takeaway: evaluate a scaffold on the decisions it produces, not the reasoning text it generates.
More from coding & agent
- Gemini CLI PR adds custom OTLP headers to telemetry, enabling auth'd exports to Grafana and Datadog — jesussamuel-byte · 2026-10-06
- A 4-step Opus 5.5 + Hermes workflow that turns competitor pages and reviews into testable positioning angles — VibeMarketer_ · 2026-10-06
- Jeff Weinstein courts SaaS platforms to embed agent-driven purchasing flows with approval gates — jeff_weinstein · 2026-10-06
- Is Product the new DevRel? DevRel teams are merging into product and PM roles — gethackteam · 2026-10-06
- How 4 humans coordinate Claude Code, Codex, Muse and Hermes via Microsoft Planner kanban — DJAI9LAB · 2026-10-06
- HyperFrames Studio launches: a video editor built for agents, co-editing on one timeline — toolstelegraph · 2026-10-06