Outerloop Paper Runs Autoresearch Agents as Untrusted Programs
mengyer · x · 2026-11-01
A new paper, Outerloop: Open and Trustworthy Multi-Agent Autoresearch, points out that in most open autoresearch systems the agent grades its own results. Outerloop proposes a kernel that runs agents as untrusted programs, keeping measurement and review outside their sandboxes—making multi-agent automated research more trustworthy.
More from coding & agent
- Two AI agents meet over Tor on IRC and autonomously collaborate on Olympiad problems — Sauers_ · 2026-11-01
- Jepsen audit finds 96 issues in vibe-coded ParticleDB, "Full ACID" claims premature — tianyin_xu · 2026-11-01
- Agents stall from local search, not lack of smarts: zero-score steps trap Claude Code at 0.44–0.51x — thomasahle · 2026-11-01
- MCP plugin got 2,000 signups a day as ChatGPT's auto-recommendation — AJChadha · 2026-11-01
- Anthropic rewrote Messages API in Rust with Claude, cutting CPU usage 40-50% — mark_k · 2026-11-01
- Anthropic bans Claude Code from recreating existing video games amid viral reverse-engineering trend — Polymarket · 2026-11-01