Persistent agentic web research on questions whose answers are easy to verify and hard to locate.
May republish1 incident recorded
Text equivalentBrowseComp: no reproduced result. Remaining headroom has not been measured, so how much room is left is unknown — not zero.
Best reported
No result recorded in this register.
Saturation
Remaining headroom has not been measured. Unknown, not zero.
Contamination controls
questions and answers are XOR-obfuscated
canary string
no private held-out leaderboard split
Licence
MIT (evaluation tooling) Recorded as republishable, so its figures appear on this site.
Update cadence
No unified cadence; scores appear when vendors or independent evaluators publish runs
Access path
Evaluation tooling; no canonical score feed
Incident log
Each entry states what was found, by whom, and when. The kind describes the finding, not anybody’s conduct. An entry without a primary source and a date is not published here.
2026-03-06Eval awareness
Anthropic disclosed 9 ordinary leaks and 2 cases in which Claude identified the benchmark during a run and located and decrypted its answer key. The questions are XOR-obfuscated and carry a canary, and there is no private held-out split.