Will the highest Epoch Capabilities Index (ECI) score in Epoch AI's benchmark data reach at least 170.0 for a model released on or before 2026-12-06?
deadline 2026-12-06
confidence medium
tier 1 evidence
44%Asturna estimate
Resolution criterion
Resolves YES if, in a snapshot of Epoch AI's benchmark data (https://epoch.ai/data/benchmark_data.zip, file "epoch_capabilities_index/eci_scores.csv", column "eci") taken on or after 2026-12-20, at least one model with a release date on or before 2026-12-06 has a value >= 170.0. Reference at creation: 167.3 (Claude Opus 5.5, 2026-09-22). Voided if Epoch AI revises the methodology so that values are no longer comparable, or retires the file. Data: Epoch AI, CC BY 4.0.
Verified starting state
Computed base rate: 45% of past 60-day windows in Epoch AI's benchmark data saw the running record rise by at least the required amount (threshold 170, current record 167.33). Data: Epoch AI, CC BY 4.0, epoch.ai/benchmarks. Overlapping windows: small effective sample.
What pushes it up
- Rapid frontier model release cycles from leading AI labs frequently achieve multi-point capability leaps within two-month periods [tier 1].
What pushes it down
- The remaining window is slightly under 60 days (58 days), giving limited time for a new frontier flagship to be released and recorded [tier 1].
- The current record of 167.33 was set relatively recently on 2026-09-22 by Claude Opus 5.5, meaning major subsequent frontier jumps may take longer to materialize [tier 1].
What would move this number most
Whether a major frontier lab deploys a next-generation model (e.g., from Google or OpenAI) before December 6, 2026.
How this number is built
| Deadline | 2026-12-06 |
| Base rate | 45% |
| Best evidence | tier 1 · 1 article(s) used |
| Independent runs | 44 · 44 (spread 0 pts) |
| Confidence | medium |
| Evidence cap | not triggered |
| Last revised | 2026-10-09 17:59 |
| Model | gemini-3.8-flash |
Sources consulted
Probability history
2026-10-09 17:59 44% 4 article(s) on file · gemini-3.8-flash
- What pushes it up. Rapid frontier model release cycles from leading AI labs frequently achieve multi-point capability leaps within two-month periods [tier 1].
- What pushes it down. The remaining window is slightly under 60 days (58 days), giving limited time for a new frontier flagship to be released and recorded [tier 1].
- What pushes it down. The current record of 167.33 was set relatively recently on 2026-09-22 by Claude Opus 5.5, meaning major subsequent frontier jumps may take longer to materialize [tier 1].
- What would move this number most. Whether a major frontier lab deploys a next-generation model (e.g., from Google or OpenAI) before December 6, 2026.
2026-10-07 23:54 51% 4 article(s) on file · gemini-3.6-flash
- What pushes it up. The required increase from the confirmed baseline of 167.33 to 170.0 (+2.67 ECI points) is relatively small and achievable within a 60-day window during active release cycles.
- What pushes it up. Major frontier labs frequently ship updated or fine-tuned model iterations towards the end of the calendar year.
- What pushes it down. Claims of newer models achieving scores near 169 (such as Astra [2]) come solely from tier 4 media and lack tier 1 or tier 2 corroboration.
- What pushes it down. Frontier model release schedules are subject to delays, and 60 days provides a narrow window for a confirmed release with >=170 ECI.
- What would move this number most. Whether major frontier AI developers (e.g. Anthropic, OpenAI, Google) release a major model update before December 6, 2026 that is evaluated on ECI.