Part of event: Highest Claude score on Humanity’s Last Exam in 2026? · highest-claude-score-on-humanitys-last-exam-in-2026-20260723190836285
Market structure and 24h change as reported by Polymarket's public API.
Top of book
Best bid
0.845
Best ask
0.860
Spread
0.015
As reported by Polymarket's public API at 12:10 UTC; top of book, not full depth; per-side depth not shown.
Volume
24h volume
$130
Total volume
$7,400
Volume as reported by Polymarket's public API.
Resolution rules
Rules as stated by Polymarket, reproduced verbatim. Convexly has not independently reviewed them and does not endorse them; the rating's resolution-reliability pillar treats rule text as present but not independently reviewed.
This market will resolve to "Yes" if any Anthropic Claude model achieves at least the specified accuracy on Humanity’s Last Exam by December 31, 2026, 11:59 PM ET. Otherwise, this market will resolve to "No". For resolution, “accuracy” refers solely to the value labeled “HLE Accuracy” or a clear equivalent metric if the data’s presentation or terminology is restructured, regardless of the model’s Calibration Error or any other displayed metric. The resolution source will be the official Humanity’s Last Exam leaderboard at https://agi.safe.ai/. If the resolution source becomes unavailable during the listed timeframe, this market will remain open to allow the relevant data to become available again. If the source remains unavailable after the end of the listed timeframe or is otherwise confirmed to be permanently unavailable, official Humanity’s Last Exam results published elsewhere may be used. If no official alternative source is available, this market will resolve to "No".Show the full rules textHide the full rules text
This market will resolve to "Yes" if any Anthropic Claude model achieves at least the specified accuracy on Humanity’s Last Exam by December 31, 2026, 11:59 PM ET. Otherwise, this market will resolve to "No". For resolution, “accuracy” refers solely to the value labeled “HLE Accuracy” or a clear equivalent metric if the data’s presentation or terminology is restructured, regardless of the model’s Calibration Error or any other displayed metric. The resolution source will be the official Humanity’s Last Exam leaderboard at https://agi.safe.ai/. If the resolution source becomes unavailable during the listed timeframe, this market will remain open to allow the relevant data to become available again. If the source remains unavailable after the end of the listed timeframe or is otherwise confirmed to be permanently unavailable, official Humanity’s Last Exam results published elsewhere may be used. If no official alternative source is available, this market will resolve to "No".
Resolution mechanics
Resolution parameters as stated by Polymarket.
- Resolved by (UMA adapter)
- 0x65070BE91477460D8A7AeEb94ef92fe056C2f2A7
- Proposer bond
- $500 proposer bond
- Proposer reward
- $4 proposer reward
- Stated resolution source
- https://agi.safe.ai/
Market Trust diagnostic · v0.2 candidate
Will the highest score achieved by an Anthropic Claude model on Humanity’s Last Exam in 2026 be 55% or higher?
This market is not in the v0.1 daily card set yet. The v0.2 diagnostic scorecard below shows the per-pillar contribution from the latest materializer run, with visible caveats and evidence-status separation.
Why this rating
Why this rating
Composite 44 · Do not cite
Measurement certainty
Low
2 measured · 3 modeled · 1 not assessed
Certainty reflects how much of this market we measured, not how good the market is.
- CoherenceMeasuredlow readingweight 25.0
4 full-provenance Coherent Markets Engine signals linked to this condition; 22 price-observation checkpoints recorded.
- · full provenance cme signal linked
- · signal price observations linked
- · no observed checkpoint yet
30+9.4 - ExecutabilityModeled estimatefixed rules, not a per-market measurementweight 20.0
Executability scored as worse-of(spread, depth) from a single orderbook snapshot; the two were merged from the former liquidity + depth pillars to stop double-counting one correlated source.
- · executability worse of spread depth
- · spread: top-of-book spread 1.5c
- · spread: spread is 2% of mid, tight relative to price
- · spread: pp-band-to-cents-band approximation in effect (depth_5pct_usd/2 per side); per-side decomposition pending substrate work
15+3.8 - Resolution reliabilityModeled estimatefixed rules, not a per-market measurementweight 20.0
Resolution source and rule text are present but not independently reviewed.
- · resolution metadata present manual review pending
70+17.5 - Participant flowNot assessednot yet measured for this marketweight 20.0
Participant flow inputs exist but do not meet identity-coverage floors; descriptive only.
- · below publishable sample floor
- · high participant concentration
- · high unknown flow share
- · insufficient scored accounts
not assessednot assessed - Integrity risk screenModeled estimatefixed rules, not a per-market measurementweight 10.0
Heavy concentration / wash-pattern signals; treat with caution.
- · top-3 wallets control 100% of notional flow
- · 67% of wallet/side pairs appear repeatedly
- · flow events: 11
- · distinct wallets: 3
55+6.9 - Audit completenessMeasuredweight 5.0
3 linked artifacts; 0 missing hashes; 0 stale/failing sources.
- · artifact hashes present
- · source watermarks healthy
100+6.3
How it adds up: composite = Σ(score × weight) / Σ(weight) across measured + heuristic pillars only. Pending and data-rights-blocked pillars are excluded from both numerator and denominator. Total contributing weight = 80.0.
Caveat: 1 pillar not assessed yet; the score will recompute as the substrate fills in.
Structural coherence read
The Coherent Markets Engine checks this market's pricing for structural inconsistencies against related markets. Whether a current signal exists, and its construction, is included with Researcher.
See ResearcherMarket opened 2026-07-23 · minimum order size 5 shares · price tick size 0.001 · as stated by Polymarket.