Understand this story
On October 17, 2026, Hao AI Lab announced the results of its multi-agent crime-solving benchmark, which used 200 Victorian London detective cases. Among 100 harder cases in which paid witnesses outnumbered honest ones, GPT6-Astra leading the Jev cluster solved 93 cases, GPT6-Astra alone solved 59 cases, the Jev cluster alone solved 42 cases, and a rule-based leader plus the Jev cluster solved only 0–1 cases. Hao AI Lab also noted that when working alone, each of the two approaches has its own failure mode: the leaderless Jev swarm is like a rumor mill—one child misreads a clue, its companions copy the vote, and the whole street accepts the wrong answer; while the lone genius is limited by time and cannot deliver the decisive clue to the door. The authors conclude that, in the face of organized lies, a reasoning leader is needed who will ask, "Who is speaking?"
AI-generated from public reports
Latest development · Report
Hao AI Lab: GPT6-Astra-led Jev agent swarm murder mystery experimentDevelopments
1 confirmed developments
Hao AI Lab announces results of multi-agent crime-solving evaluation
Source:UCSD Hao AI Lab
Reports and discussion timeline
Browse reports and discussion by time. Their evidence levels remain distinct.
Oct 17
UCSD Hao AI LabFirst-party
Hao AI Lab: GPT6-Astra-led Jev agent swarm murder mystery experiment
Hao AI Lab tested agent swarms on 200 Victorian London murder cases: the Jev agent swarm alone, GPT6-Astra alone, and GPT6-Astra leading the Jev swarm. The conclusion: let fast agents run errands and strong reasoners think—when facing organized lies, you need a reasoning leader to judge who is speaking.
UCSD Hao AI LabFirst-party
Hao AI Lab: Leaderless Swarms and Lone Wolves Each Fail in Their Own Way
Hao AI Lab points out in its tweet that when operating alone, each of the two approaches fails in its own way: the leaderless Jev swarm is like a rumor mill—one child misreads a clue, their companions copy the vote, and the whole street ends up agreeing on the wrong answer; while the lone genius wears out their shoes, as a single taxi can never deliver the decisive clue to the door within the limited time.
UCSD Hao AI LabFirst-party
Hao AI Lab: GPT6-Astra as Leader in the Jev Cluster Crime-Solving Test
Hao AI Lab tested agent clusters on 200 Victorian London detective cases: in 100 harder cases where paid witnesses outnumbered honest ones, GPT6-Astra leading the Jev cluster solved 93 cases, GPT6-Astra alone solved 59, the Jev cluster alone solved 42, and a rule-based leader plus the Jev cluster solved only 0–1. The authors conclude that, in the face of organized lies, a reasoning leader is needed who will keep asking "who is speaking."