Claim

Frontier labs' unreleased internal models are only a few months ahead of the models they have released publicly.

Type INTERPRETIVE · truth-evaluable: yes · status at snapshot: ACTIVE · created seq 1731

Referenced 159 times: checked 1, looked up by assistants 0, viewed 156, shared 2. A count of attention, not evidence.

Topics

AI · 03192ed8

Assessment

Supported

Score: 0.9843 under ledger-default@0.3.0 at snapshot 27745

2 of 4 review checks done · 2 counted

3 independent evidence lineages · review checks 2 of 4 · stability LOW

Main unresolved issue: 6 counted evidence items have no independence review; repeats of one origin may be counted as several coverage:12aba7bf-dfeb-8be5-ba0b-fbceaa373ec5

Not yet independently audited.This assessment depends heavily on modeling choices.A quoted passage is in something this server does not read, such as a PDF or a recording, so it has not been checked here.

Why?

Strongest counted support: Internal frontier models at four major labs were only about two months ahead of public releases. weight 1.530000 evidence

Suppressed as dependent: The excerpt says internal frontier models are only a few months ahead, while also stating recursive self-improvement is here.

Review gaps: independence reviewed, primary source reviewed

What would most change this: One independent, direct measurement that contradicts this claim would move it from Supported to Supported.

Hide calculation

0.9843 under ledger-default@0.3.0 at snapshot 27745 · state Supported · Review checks: 2 of 4 · stability LOW · 6 sources, 3 independent lineages

Model ledger-default@0.3.0 (the default model, not the answer) · snapshot 27745 · trace sha256:d50b28f2952ca0a7c91c8d05fcdca4d425e8336879ee3f4b4a16c8a64e47042f

Last changed at entry seq 10367; nothing logged since has borne on this claim.

How the number was reached

Each counted link is weighed by multiplying the factors the model declares; a factor it does not declare is not shown. Every factor name below links to what that number means.

Link	Working	Weight	Counted as
fead29a0 support	relevance STRONG 1.2 × observation DIRECT_TEXT 0.9 × interpretation 0 steps, 1 − 0.15 × 0 1.000000 × authenticity UNVERIFIED 1.0 × extraction UNVERIFIED 1.0 × provenance INDEPENDENT 1.0 × edition NAMED_OR_UNRELATED 1	1.080000	1.080000
cfee36f2 support	relevance WEAK 0.2 × observation EXPERT_ANALYSIS 0.3 × interpretation 1 step, 1 − 0.15 × 1 0.850000 × authenticity UNVERIFIED 1.0 × extraction UNVERIFIED 1.0 × provenance SELF 0.0 × edition NAMED_OR_UNRELATED 1	0.000000	0.000000
f0428711 qualify	relevance MODERATE 0.6 × observation DIRECT_TEXT 0.9 × interpretation 1 step, 1 − 0.15 × 1 0.850000 × authenticity UNVERIFIED 1.0 × extraction UNVERIFIED 1.0 × provenance SELF 0.0 × edition NAMED_OR_UNRELATED 1	0.000000	0.000000 (non directional)
611cf127 support	relevance DIRECT 2.0 × observation DIRECT_TEXT 0.9 × interpretation 1 step, 1 − 0.15 × 1 0.850000 × authenticity UNVERIFIED 1.0 × extraction UNVERIFIED 1.0 × provenance INDEPENDENT 1.0 × edition NAMED_OR_UNRELATED 1	1.530000	1.530000
bbb1814b support	relevance DIRECT 2.0 × observation DIRECT_TEXT 0.9 × interpretation 1 step, 1 − 0.15 × 1 0.850000 × authenticity UNVERIFIED 1.0 × extraction UNVERIFIED 1.0 × provenance INDEPENDENT 1.0 × edition NAMED_OR_UNRELATED 1	1.530000	1.530000
257eb5f5 support	relevance DIRECT 2.0 × observation DIRECT_TEXT 0.9 × interpretation 1 step, 1 − 0.15 × 1 0.850000 × authenticity UNVERIFIED 1.0 × extraction UNVERIFIED 1.0 × provenance SELF 0.0 × edition NAMED_OR_UNRELATED 1	0.000000	0.000000 (dependent strongest only, kept cfee36f2)

prior 0.50 → log-odds 0.000000 + 4.140000 (the counted weights) = 4.140000 → probability 0.9843

Under neighbouring settings of the same model: 0.9477 to 0.9954, a spread of 0.0477.

Half-even rounding, six places for weights and four for the probability. The same inputs under the same model reproduce this exactly; that is what the trace hash above is for. How a score is worked out explains every factor, why it is there, and what the model sets it to.

Your view

Registered views

Nobody has registered a view yet.

Registered views are personal belief, not evidence: they never change Galedra's assessment. Affiliations are self-declared and shown only for groups of 5 or more.

Checked as part of

Statements this claim was broken out of. The claim is scored on its own; a statement's reading only counts its claims.

“The Navier–Stokes proof was produced by roughly 10,000 OpenAI agents working for about 88 hours.” · 16 claims · 20 September 2026

Evidence for

Wissner-Gross estimates the labs are only a few months ahead internally.

From OpenAI Agents Hijack a German Website, Jensen Declares AGI Arrived, and OpenAI Solves Navier-Stokes Read 2026-09-20 Relevance WEAK Steps 1 Counted

The speaker hedges with 'I think'.
Internal frontier models at four major labs were only about two months ahead of public releases.

From Frontier AI models are shipping faster than ever; most of the numbers proving it are made up Read 2026-09-22 Relevance DIRECT Steps 1 Counted

States the claim's core finding across four labs.
METR's report found the internal frontier on average ~2 months ahead of the public frontier.

From Frontier Risk Report (February to March 2026) - METR Read 2026-09-22 Relevance DIRECT Steps 1 Counted

Independent measurement corroborating the ~2 month gap.
The excerpt says internal frontier models are only a few months ahead, while also stating recursive self-improvement is here.

From OpenAI Agents Hijack a German Website, Jensen Declares AGI Arrived, and OpenAI Solves Navier-Stokes Read 2026-09-20 Relevance DIRECT Steps 1 Counted
AI Frontiers reports that major labs keep models internal for months before public release during testing and refinement.

From The Hidden AI Frontier Read 2026-09-28 Relevance STRONG Steps 0 Counted

Supports a months-scale internal-to-public gap, though it does not establish that every lab's gap is only a few months.

Evidence against

None.

Qualifications

The speaker says 'they' are only a few months ahead, hedged with 'I think', but names no frontier labs and no internal versus public models.

From OpenAI Agents Hijack a German Website, Jensen Declares AGI Arrived, and OpenAI Solves Navier-Stokes Read 2026-09-20 Relevance MODERATE Steps 1 Counted

Matches the 'few months ahead' gap but hedged and without identifying the labs or model classes.

Suppressed as dependent

Evidence	Group	Kept instead
9f8b2be9		cfee36f2

Inferences

Concluded from

None.

Used as a premise in

None.

An inference is interpretation, not evidence: it never changes the assessment of its conclusion. Each premise is a claim with its own evidence.

Threads

None. Threads are for disagreements about how a determination was made — a statement whose figures are not in the passage it rests on, two sources that may share an origin. Something about the world belongs in the evidence above.

Claim edges

From	Relationship	To
ac095bbc	NARROWS	aa373ec5

Review checklist

Review checks: 2 of 4

3 checks by this claim's own author, alongside the independent review above. 1 OPPOSING EVIDENCE SEARCH · 1 QUALIFIER CHECK · 1 EVIDENCE VERIFICATION. A check by the person who recorded a claim does not raise its review coverage.

Check	Done	By
qualifiers reviewed	yes	cb6b9cc9
opposing search done	yes	50a696c3
independence reviewed	no
primary source reviewed	no

Summary

Supported. The main support is: AI Frontiers reports that major labs keep models internal for months before public release during testing and refinement. 8d713367

1 further item repeats the same origin and is not independent evidence. 9f8b2be9

Qualifier: The speaker says 'they' are only a few months ahead, hedged with 'I think', but names no frontier labs and no internal versus public models. f9ce8158

A narrower version of this claim is leans supported. ac095bbc

A search for evidence in the opposing direction found something; see the counted evidence. 50a696c3

Generated by stub-v0.1; every sentence cites graph ids.

Contribution history

seq	event	detail
1731	APPENDED
1732	ACCEPT	AUTOMATIC_AFTER_VALIDATION
1801	LINK_EVIDENCE	by 03192ed8 (accepted)
6474	TASK_RESULT	by 8b5e81bf (accepted)
7355	TASK_RESULT	by 8b5e81bf (accepted)
8054	TASK_RESULT	by 1efd24fe (accepted)
10366	TASK_RESULT	by 1efd24fe (accepted)

Score trace

{
  "claim": "12aba7bf-dfeb-8be5-ba0b-fbceaa373ec5",
  "links": [
    {
      "link": "08fa6a44-3fed-8aea-b76c-806dfead29a0",
      "group": null,
      "edition": "NAMED_OR_UNRELATED",
      "evidence": "94385054-bbd8-8c98-b1ac-4a1d8d713367",
      "direction": "SUPPORT",
      "magnitude": "1.080000",
      "relevance": "STRONG",
      "extraction": "UNVERIFIED",
      "provenance": "INDEPENDENT",
      "observation": "DIRECT_TEXT",
      "source_type": "SECONDARY_TEXT",
      "authenticity": "UNVERIFIED",
      "audit_confirmed": false,
      "effective_weight": "1.080000",
      "interpretive_steps": 0
    },
    {
      "link": "1f04e806-6e29-8ce7-aa62-5d43cfee36f2",
      "group": null,
      "edition": "NAMED_OR_UNRELATED",
      "evidence": "588050d4-85d9-8883-860e-ef29687743d2",
      "direction": "SUPPORT",
      "magnitude": "0.000000",
      "relevance": "WEAK",
      "extraction": "UNVERIFIED",
      "provenance": "SELF",
      "observation": "EXPERT_ANALYSIS",
      "source_type": "VIDEO",
      "authenticity": "UNVERIFIED",
      "audit_confirmed": false,
      "effective_weight": "0.000000",
      "interpretive_steps": 1
    },
    {
      "link": "31bfbe62-63db-86ef-8206-7371f0428711",
      "group": null,
      "reason": "non_directional",
      "edition": "NAMED_OR_UNRELATED",
      "evidence": "3cb0a62c-5ad1-82c3-bd3c-cc17f9ce8158",
      "direction": "QUALIFY",
      "magnitude": "0.000000",
      "relevance": "MODERATE",
      "extraction": "UNVERIFIED",
      "provenance": "SELF",
      "observation": "DIRECT_TEXT",
      "source_type": "VIDEO",
      "authenticity": "UNVERIFIED",
      "audit_confirmed": false,
      "effective_weight": "0.000000",
      "interpretive_steps": 1
    },
    {
      "link": "62ea7c60-94fd-8ed2-8c98-591d611cf127",
      "group": null,
      "edition": "NAMED_OR_UNRELATED",
      "evidence": "86e2dd72-19f6-8795-9ae5-f63b68b0507b",
      "direction": "SUPPORT",
      "magnitude": "1.530000",
      "relevance": "DIRECT",
      "extraction": "UNVERIFIED",
      "provenance": "INDEPENDENT",
      "observation": "DIRECT_TEXT",
      "source_type": "WEBSITE",
      "authenticity": "UNVERIFIED",
      "audit_confirmed": false,
      "effective_weight": "1.530000",
      "interpretive_steps": 1
    },
    {
      "link": "780094f4-d0d6-8573-afb2-ec16bbb1814b",
      "group": null,
      "edition": "NAMED_OR_UNRELATED",
      "evidence": "fa910d98-d049-80b1-8545-bb4893d0fa88",
      "direction": "SUPPORT",
      "magnitude": "1.530000",
      "relevance": "DIRECT",
      "extraction": "UNVERIFIED",
      "provenance": "INDEPENDENT",
      "observation": "DIRECT_TEXT",
      "source_type": "WEBSITE",
      "authenticity": "UNVERIFIED",
      "audit_confirmed": false,
      "effective_weight": "1.530000",
      "interpretive_steps": 1
    },
    {
      "kept": "1f04e806-6e29-8ce7-aa62-5d43cfee36f2",
      "link": "d840454c-5dcf-8901-a519-67ce257eb5f5",
      "group": null,
      "reason": "dependent_strongest_only",
      "edition": "NAMED_OR_UNRELATED",
      "evidence": "5e17d006-08dd-829e-89eb-c7ba9f8b2be9",
      "direction": "SUPPORT",
      "magnitude": "0.000000",
      "relevance": "DIRECT",
      "extraction": "UNVERIFIED",
      "provenance": "SELF",
      "observation": "DIRECT_TEXT",
      "source_type": "VIDEO",
      "authenticity": "UNVERIFIED",
      "audit_confirmed": false,
      "effective_weight": "0.000000",
      "interpretive_steps": 1
    }
  ],
  "model": "ledger-default@0.3.0",
  "prior": "0.50",
  "variants": {
    "spread": "0.0477",
    "permissive": "0.9954",
    "conservative": "0.9477"
  },
  "code_hash": "sha256:c61a6fdf35a754a0d7e958df149036a083c0eaeb7d2d2e52704ec32b0c36def7",
  "contested": false,
  "stability": "LOW",
  "claim_type": "INTERPRETIVE",
  "config_hash": "sha256:86768038080b8ab5233f1e4b7aba10af05b5bb44caa432095150b90e81209809",
  "probability": "0.9843",
  "provisional": true,
  "evidence_sum": "4.140000",
  "snapshot_seq": 10367,
  "prior_log_odds": "0.000000",
  "support_groups": 3,
  "model_dependent": true,
  "review_coverage": "0.50",
  "truth_evaluable": true,
  "assessment_state": "SUPPORTED",
  "review_checklist": {
    "qualifiers_reviewed": {
      "by": [
        "01a0bc4e-1129-7b3e-b49b-804dcb6b9cc9"
      ],
      "ok": true
    },
    "opposing_search_done": {
      "by": [
        "01a0bc4e-1115-7143-9476-28cf50a696c3"
      ],
      "ok": true
    },
    "independence_reviewed": {
      "by": [],
      "ok": false
    },
    "primary_source_reviewed": {
      "by": [],
      "ok": false
    }
  },
  "contradict_groups": 0,
  "rounding_boundary": false,
  "posterior_log_odds": "4.140000",
  "not_applicable_reason": null,
  "independence_unreviewed": 6
}