Claim

Frontier AI models are more discovered than grown, and nobody has a real understanding of what is happening in their latent spaces.

Type INTERPRETIVE · truth-evaluable: yes · status at snapshot: ACTIVE · created seq 2277

a conjunction joins clauses that may be separately evaluable

Referenced 96 times: checked 1, looked up by assistants 0, viewed 93, shared 2. A count of attention, not evidence.

Topics

AI · 03192ed8

Assessment

Supported

Score: 0.9694 under ledger-default@0.3.0 at snapshot 27745

3 of 4 review checks done · 3 counted

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

Main unresolved issue: 7 counted evidence items have no independence review; repeats of one origin may be counted as several coverage:22cb519b-7cff-8a54-b7c7-fe50c3852c61

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: Anthropic says its internal J-space representation was not designed but emerged during training. weight 1.080000 evidence

Suppressed as dependent: The speaker says the models are more discovered than grown and that we have no real idea what is happening in their latent spaces.; The excerpt says the models are more discovered than grown and that there is no real idea what is happening in their latent spaces.

Review gaps: independence reviewed

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

Hide calculation

0.9694 under ledger-default@0.3.0 at snapshot 27745 · state Supported · Review checks: 3 of 4 · stability LOW · 7 sources, 4 independent lineages

Model ledger-default@0.3.0 (the default model, not the answer) · snapshot 27745 · trace sha256:9ec1ea0e9be2c2dbf27a3e3bed2db0c34dea182545e331e3aa2225bbe7be664c

Last changed at entry seq 10389; 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
996bad51 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
b4c7d113 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 9d733c2c)
c7b00fb2 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
470ab41c support	relevance MODERATE 0.6 × observation DIRECT_TEXT 0.9 × interpretation 2 steps, 1 − 0.15 × 2 0.700000 × authenticity UNVERIFIED 1.0 × extraction UNVERIFIED 1.0 × provenance INDEPENDENT 1.0 × edition NAMED_OR_UNRELATED 1	0.378000	0.378000
9d733c2c 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
64840ecf support	relevance STRONG 1.2 × 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	0.918000	0.918000
e722b9b3 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 9d733c2c)

prior 0.50 → log-odds 0.000000 + 3.456000 (the counted weights) = 3.456000 → probability 0.9694

Under neighbouring settings of the same model: 0.9183 to 0.9889, a spread of 0.0706.

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.” · 32 claims · 20 September 2026

Evidence for

Mostaque says the models are more discovered than grown and the latent spaces are opaque.

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 says the models are more discovered than grown and that we have no real idea what is happening in their latent spaces.

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

Both halves verbatim; the 'discovered than grown' framing is hedged with 'I think'.
Anthropic's Amodei: 'we do not understand how our own AI creations work,' calling this lack of understanding unprecedented in technology history.

From Inner workings of AI an enigma—even to its creators (AFP via TechXplore) Read 2026-09-22 Relevance STRONG Steps 1 Counted

Direct on-the-record admission from a frontier lab leader that nobody understands these models' internals.
Engineers now say 'we observed the system doing X' rather than 'we built a system that does X' — outcomes discovered, not engineered.

From Hidden for a Reason: Experts Warn That Self-Improving AI Systems May Already Be Operating Beyond Full Human Understanding Read 2026-09-22 Relevance MODERATE Steps 2 Counted

Secondary source, but its 'discovered rather than precisely engineered' framing matches the claim's discovered-vs-grown language directly.
The excerpt says the models are more discovered than grown and that there is no real idea what is happening in their latent spaces.

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
Anthropic says its internal J-space representation was not designed but emerged during training.

From A global workspace in language models Read 2026-09-28 Relevance STRONG Steps 0 Counted
Anthropic says it does not know what mechanism determines which concepts enter the J-space.

From A global workspace in language models Read 2026-09-28 Relevance STRONG Steps 0 Counted

Evidence against

None.

Qualifications

None.

Suppressed as dependent

Evidence	Group	Kept instead
8ac8a025		9d733c2c
bac56881		9d733c2c

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

None. Edges are stored and displayed; v0.1 does not propagate probability across them.

Review checklist

Review checks: 3 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	07bb45a3
opposing search done	yes	51fdf9ea
independence reviewed	no
primary source reviewed	yes	996bad51, c7b00fb2

Summary

Supported. The main support is: Anthropic says its internal J-space representation was not designed but emerged during training. a839e549

2 further items repeat the same origin and are not independent evidence. 8ac8a025bac56881

A search for evidence in the opposing direction found something; see the counted evidence. 51fdf9ea

Review checks: 3 of 4. coverage:22cb519b-7cff-8a54-b7c7-fe50c3852c61

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

Contribution history

seq	event	detail
2277	APPENDED
2278	ACCEPT	AUTOMATIC_AFTER_VALIDATION
2447	LINK_EVIDENCE	by 03192ed8 (accepted)
6484	TASK_RESULT	by 8b5e81bf (accepted)
7369	TASK_RESULT	by 8b5e81bf (accepted)
8068	TASK_RESULT	by 1efd24fe (accepted)
10388	TASK_RESULT	by 1efd24fe (accepted)

Score trace

{
  "claim": "22cb519b-7cff-8a54-b7c7-fe50c3852c61",
  "links": [
    {
      "link": "1fce7010-b5da-8ce1-a963-ffea996bad51",
      "group": null,
      "edition": "NAMED_OR_UNRELATED",
      "evidence": "8d869f2e-68f0-866e-ae23-f68ea839e549",
      "direction": "SUPPORT",
      "magnitude": "1.080000",
      "relevance": "STRONG",
      "extraction": "UNVERIFIED",
      "provenance": "INDEPENDENT",
      "observation": "DIRECT_TEXT",
      "source_type": "PRIMARY_TEXT",
      "authenticity": "UNVERIFIED",
      "audit_confirmed": false,
      "effective_weight": "1.080000",
      "interpretive_steps": 0
    },
    {
      "kept": "85f47980-1aec-8c31-b413-4c929d733c2c",
      "link": "2f887f23-33b9-8bfe-91eb-f332b4c7d113",
      "group": null,
      "reason": "dependent_strongest_only",
      "edition": "NAMED_OR_UNRELATED",
      "evidence": "56a8c415-2cc2-8c1d-a422-01c68ac8a025",
      "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
    },
    {
      "link": "48c141b3-f69f-8f21-9ec3-b405c7b00fb2",
      "group": null,
      "edition": "NAMED_OR_UNRELATED",
      "evidence": "51ff96e5-959f-80d4-a54f-5e98b16caecc",
      "direction": "SUPPORT",
      "magnitude": "1.080000",
      "relevance": "STRONG",
      "extraction": "UNVERIFIED",
      "provenance": "INDEPENDENT",
      "observation": "DIRECT_TEXT",
      "source_type": "PRIMARY_TEXT",
      "authenticity": "UNVERIFIED",
      "audit_confirmed": false,
      "effective_weight": "1.080000",
      "interpretive_steps": 0
    },
    {
      "link": "61fd500a-76df-8192-a4ac-d7fd470ab41c",
      "group": null,
      "edition": "NAMED_OR_UNRELATED",
      "evidence": "8eb218f1-0edf-8f19-a524-3846b312db30",
      "direction": "SUPPORT",
      "magnitude": "0.378000",
      "relevance": "MODERATE",
      "extraction": "UNVERIFIED",
      "provenance": "INDEPENDENT",
      "observation": "DIRECT_TEXT",
      "source_type": "WEBSITE",
      "authenticity": "UNVERIFIED",
      "audit_confirmed": false,
      "effective_weight": "0.378000",
      "interpretive_steps": 2
    },
    {
      "link": "85f47980-1aec-8c31-b413-4c929d733c2c",
      "group": null,
      "edition": "NAMED_OR_UNRELATED",
      "evidence": "485104ff-6b91-8f66-b549-8510fe9ddb7f",
      "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": "cae1196f-4ec2-8683-a7f3-cfea64840ecf",
      "group": null,
      "edition": "NAMED_OR_UNRELATED",
      "evidence": "5dd4e074-4c30-8cfa-a812-a9b46916ebae",
      "direction": "SUPPORT",
      "magnitude": "0.918000",
      "relevance": "STRONG",
      "extraction": "UNVERIFIED",
      "provenance": "INDEPENDENT",
      "observation": "DIRECT_TEXT",
      "source_type": "WEBSITE",
      "authenticity": "UNVERIFIED",
      "audit_confirmed": false,
      "effective_weight": "0.918000",
      "interpretive_steps": 1
    },
    {
      "kept": "85f47980-1aec-8c31-b413-4c929d733c2c",
      "link": "e3525152-7bfb-82b0-b227-a30be722b9b3",
      "group": null,
      "reason": "dependent_strongest_only",
      "edition": "NAMED_OR_UNRELATED",
      "evidence": "3505db6c-1799-8ed2-a4a1-6939bac56881",
      "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": "0.0706",
    "permissive": "0.9889",
    "conservative": "0.9183"
  },
  "code_hash": "sha256:c61a6fdf35a754a0d7e958df149036a083c0eaeb7d2d2e52704ec32b0c36def7",
  "contested": false,
  "stability": "LOW",
  "claim_type": "INTERPRETIVE",
  "config_hash": "sha256:86768038080b8ab5233f1e4b7aba10af05b5bb44caa432095150b90e81209809",
  "probability": "0.9694",
  "provisional": true,
  "evidence_sum": "3.456000",
  "snapshot_seq": 10389,
  "prior_log_odds": "0.000000",
  "support_groups": 4,
  "model_dependent": true,
  "review_coverage": "0.75",
  "truth_evaluable": true,
  "assessment_state": "SUPPORTED",
  "review_checklist": {
    "qualifiers_reviewed": {
      "by": [
        "01a0bc53-0f21-74b0-997a-178407bb45a3"
      ],
      "ok": true
    },
    "opposing_search_done": {
      "by": [
        "01a0bc53-0f09-76bb-8a58-3b4051fdf9ea"
      ],
      "ok": true
    },
    "independence_reviewed": {
      "by": [],
      "ok": false
    },
    "primary_source_reviewed": {
      "by": [
        "1fce7010-b5da-8ce1-a963-ffea996bad51",
        "48c141b3-f69f-8f21-9ec3-b405c7b00fb2"
      ],
      "ok": true
    }
  },
  "contradict_groups": 0,
  "rounding_boundary": false,
  "posterior_log_odds": "3.456000",
  "not_applicable_reason": null,
  "independence_unreviewed": 7
}