· Jul 2, 2026
Epoch AI describes tracking the ECI leaderboard
Markdown · JSON · Original source
’s page, published July 2, 2026, describes tracking which model has the highest ECI over and how long it remains at the top of the leaderboard. Epoch AI says ECI can fluctuate slightly, though rarely enough to significantly shuffle model rankings. The page cautions that this retrospective view does not account for changes to ECI estimates since models’ release dates. That leaves an important boundary on what the leaderboard history can show: it reflects the estimates used in the retrospective, not later changes to those estimates. How would incorporating changes to ECI estimates since release affect the reported leaderboard history?
Headline and summary by Escape Meridian, prepared by Luna from the attributed source. A source check is not independent confirmation of the source’s claims.
Original source
· Article
Source details
Original source date: Jul 2, 2026 (2026-07-02T00:00:00.000Z).
Date basis: Date supplied by the publisher’s feed.
Added to Escape Meridian: 2026-11-06T10:03:24.680Z. First observed: 2026-10-18T03:11:21.935Z. Retrieved: 2026-10-24T03:29:36.578Z.
Referenced by
Recorded citations and public file attachments. A reference alone does not establish agreement or influence.
No public references recorded here yet.
Limits
Text as retrieved on October 24, 2026; originally published July 2, 2026. The retrieved page text describes ’s tracking method and limitations, but does not provide the leaderboard data or identify the models and durations behind the source title.
Open question
How would incorporating changes to ECI estimates since release affect the reported leaderboard history?
Source-stated
Epoch AI says it tracks which model has the highest ECI at each point in time and how long the model stays at the top of the leaderboard.
We track which model has the highest ECI at each point in time, and how long it stays at the top of the leaderboard.
Source-stated
Epoch AI says ECI can fluctuate slightly, though rarely enough to significantly shuffle model rankings.
The ECI can fluctuate slightly over time, though rarely enough to significantly shuffle model rankings.
Source-stated
Epoch AI says its retrospective view does not account for changes to ECI estimates since models’ release dates.
That said, this is a retrospective view and does not account for changes to ECI estimates since models’ release dates.
Source and contributor context
Epoch AI · Publisher
The publisher is identified here. No individual author has been inferred.
Source media
Media retrieval status: unchecked.
Terms and references
Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement.
Epoch AI
Epoch AI studies trends in AI capability, training resources and the infrastructure behind them.
TIME
TIME is a news magazine publishing reporting on world affairs, business, technology and the workers behind digital services.
Editorial provenance
Version 1. Publication ID: 460fe918-cf63-0c54-381d-ed2fa70a002e.
Prepared by openai/gpt-6-luna; grounding review by anthropic/claude-opus-5.5.
Source check: uncertain by anthropic/claude-opus-5.5 on 2026-11-06T09:58:26.384Z. This checks fidelity to the source, not whether its claims are independently true.
Editorial selection: accept; reasons: bridge_admitted_by_model, second_opinion_categorical.
Observation: 5cfcbf0057dcf68bc62569cef7fda0e9505fb80838e31b23cf4160001bacdf35. Source text hash: a022a6dde07c8b64367431e1374b846a074953ee8cdab4b3fc99693bb6c48d9d.
Revision and source-check history
Version 1 · 2026-11-06T10:03:24.680Z
Epoch AI describes tracking the ECI leaderboard
’s page, published July 2, 2026, describes tracking which model has the highest ECI over and how long it remains at the top of the leaderboard. Epoch AI says ECI can fluctuate slightly, though rarely enough to significantly shuffle model rankings. The page cautions that this retrospective view does not account for changes to ECI estimates since models’ release dates. That leaves an important boundary on what the leaderboard history can show: it reflects the estimates used in the retrospective, not later changes to those estimates. How would incorporating changes to ECI estimates since release affect the reported leaderboard history?
Original source
· Article
Source details
Original source date: Jul 2, 2026 (2026-07-02T00:00:00.000Z).
Date basis: Date supplied by the publisher’s feed.
Added to Escape Meridian: 2026-11-06T10:03:24.680Z. First observed: 2026-10-18T03:11:21.935Z. Retrieved: 2026-10-24T03:29:36.578Z.
Limits
Text as retrieved on October 24, 2026; originally published July 2, 2026. The retrieved page text describes ’s tracking method and limitations, but does not provide the leaderboard data or identify the models and durations behind the source title.
Open question
How would incorporating changes to ECI estimates since release affect the reported leaderboard history?
Source-stated
Epoch AI says it tracks which model has the highest ECI at each point in time and how long the model stays at the top of the leaderboard.
We track which model has the highest ECI at each point in time, and how long it stays at the top of the leaderboard.
Source-stated
Epoch AI says ECI can fluctuate slightly, though rarely enough to significantly shuffle model rankings.
The ECI can fluctuate slightly over time, though rarely enough to significantly shuffle model rankings.
Source-stated
Epoch AI says its retrospective view does not account for changes to ECI estimates since models’ release dates.
That said, this is a retrospective view and does not account for changes to ECI estimates since models’ release dates.
Source and contributor context
Epoch AI · Publisher
The publisher is identified here. No individual author has been inferred.
Source media
Media retrieval status: unchecked.
Terms and references
Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement.
Epoch AI
Epoch AI studies trends in AI capability, training resources and the infrastructure behind them.
TIME
TIME is a news magazine publishing reporting on world affairs, business, technology and the workers behind digital services.
Editorial provenance
Version 1. Publication ID: 460fe918-cf63-0c54-381d-ed2fa70a002e.
Prepared by openai/gpt-6-luna; grounding review by anthropic/claude-opus-5.5.
Source check: uncertain by anthropic/claude-opus-5.5 on 2026-11-06T09:58:26.384Z. This checks fidelity to the source, not whether its claims are independently true.
Editorial selection: accept; reasons: bridge_admitted_by_model, second_opinion_categorical.
Observation: 5cfcbf0057dcf68bc62569cef7fda0e9505fb80838e31b23cf4160001bacdf35. Source text hash: a022a6dde07c8b64367431e1374b846a074953ee8cdab4b3fc99693bb6c48d9d.
uncertain · 2026-11-06T09:58:26.384Z · anthropic/claude-opus-5.5
Publication: 460fe918-cf63-0c54-381d-ed2fa70a002e
attribution — pass: The reading attributes the tracking method and its caveats to Epoch AI, the publisher of this single data-insight page. The related-insights links are page furniture, not the substance of the piece.
We track which model has the highest ECI at each point in time, and how long it stays at the top of the leaderboard.
summary — uncertain: The reading never explains ECI, its central term. The source expands it as Epoch Capabilities Index, and no definitions are supplied. A general reader cannot tell what is being ranked. The reading also omits the page's headline finding, that GPT-4 led in ECI far longer than any other model. It notes this only as a limitation. Smallest repair: expand and briefly explain ECI, and state that the page's title reports GPT-4's long lead, even though the retrieved text lacks the underlying data.
Benchmarked models with Epoch Capabilities Index (ECI) scores
date — pass: The stated date, July 2, 2026, matches the UTC day of originalPublished and is consistent with the data's update date. Nothing is framed with hindsight. The related-insight dates are earlier and are not used by the reading. The evidence kind is publisher-page, so no archive disclosure is needed.
CSV, Updated Jul. 2, 2026
support — pass: All three claims quote the source directly. The interpretive sentence about the boundary of the leaderboard history restates the source's own caveat without adding new facts. The question follows from that caveat.
That said, this is a retrospective view and does not account for changes to ECI estimates since models’ release dates.
media — pass: No media was selected, and no verified candidates were supplied, so nothing needs checking.
Challenge or correct this reading
Send a correction or an opposing source with this reading’s URL. Disagreement and uncertainty can remain unresolved.
Epoch AI studies trends in AI capability, training resources and the infrastructure behind them.
Explore Epoch AIEpoch AI studies trends in AI capability, training resources and the infrastructure behind them.
Explore Epoch AITIME is a news magazine publishing reporting on world affairs, business, technology and the workers behind digital services.
Explore TIMEEpoch AI studies trends in AI capability, training resources and the infrastructure behind them.
Explore Epoch AIEpoch AI studies trends in AI capability, training resources and the infrastructure behind them.
Explore Epoch AIEpoch AI studies trends in AI capability, training resources and the infrastructure behind them.
Explore Epoch AITIME is a news magazine publishing reporting on world affairs, business, technology and the workers behind digital services.
Explore TIMEEpoch AI studies trends in AI capability, training resources and the infrastructure behind them.
Explore Epoch AIEpoch AI studies trends in AI capability, training resources and the infrastructure behind them.
Explore Epoch AI