Claim Transmission Atlas · v1.0
One record.
Two clocks.
How faithfully did the press transmit the record?
Follow the distinctions through time.
Lines connect claims, articles and evidence states. They do not imply article-to-article transmission or causal ancestry.
Zoom with + / −, then use the arrow controls or drag in Pan mode. With the viewer focused, use + / −, arrow keys and Home to reset. Pan mode enables touch dragging and pinch zoom; turn it off to scroll the page normally.
The visual grammar
How to read the map
A claim trunk leads to an article twig, then a SURFACE/BODY fork, a Clock-A node and a Clock-B node. Publication dates anchor articles in time. Local spacing within each article’s branch serves legibility; it does not encode another time interval.
What was public then?
Fidelity to the record available when the article appeared, under the frozen grace rule.
What was known later?
Fidelity to the later frozen record. Fair early reporting can become stale as evidence changes.
The trunk’s warm-to-cool colour tracks the visible public-record window. July 16 marks victim disclosure / the visible-time origin, not the incident date. Article colour identifies the frozen corpus stratum; shapes and fill carry evidence state.
- General
- Business / mainstream technology
- Specialist technology / cybersecurity
Expanded 87-tree legibility export from the frozen submission assets. The complete legend is inside the figure. Selection and provenance →
The bounded result
Most distinctions survived.
Established contemporaneous material drift occurred in 7 of 83 assessable BODY observations (8.4%). Both BODY BORDERLINE cases were adjudicated to non-drift; treating both as drift gives an upper sensitivity bound of 10.8%.
| Claim family | Assessable N | Faithful | Drift | Borderline | Adjudicated | Upper bound |
|---|---|---|---|---|---|---|
| C1 populations | 20 | 16 | 2 | 2 | 10.0% | 20.0% |
| C2 motive/objective | 21 | 17 | 4 | 0 | 19.0% | 19.0% |
| C4 escape/boundary crossing | 19 | 18 | 1 | 0 | 5.3% | 5.3% |
| C5 compromise scope | 23 | 23 | 0 | 0 | 0.0% | 0.0% |
Four of seven established BODY drifts concerned motive and objective. All 23 assessable BODY C5 observations were faithful at Clock A. Eight represented BODY observations were faithful at Clock A but drift at Clock B.
Descriptive results from a bounded, outlet-weighted corpus of 24 English-language articles. Search completeness is unverified. The study does not support outlet rankings, journalist-intent claims or population-level inference.
Accessible record
Inspect the source tables
Seven frozen submission tables, reproduced directly. Column names and denominators are preserved; CSV downloads carry the original bytes.
01 · corpus
Download source CSV ↓| stratum | outlets | retained_per_outlet | stratum_retained |
|---|---|---|---|
| General | Reuters; Associated Press; BBC; The Guardian | 2 each | 8 |
| Business / Mainstream Technology | Bloomberg; Financial Times; Fortune; WIRED | 2 each | 8 |
| Specialist Technology / Cybersecurity | Ars Technica; The Register; BleepingComputer; Computer Weekly | 2 each | 8 |
02 · claim family results
Download source CSV ↓| claim_family | label | assessable_n | faithful | drift | borderline | excluded_unrepresented | excluded_novel_reporting | lower_n | lower_pct | upper_n | upper_pct | adjudicated_n | adjudicated_pct |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| C1 | C1 populations | 20 | 16 | 2 | 2 | 3 | 1 | 2 | 10.0 | 4 | 20.0 | 2 | 10.0 |
| C2 | C2 motive/objective | 21 | 17 | 4 | 0 | 2 | 1 | 4 | 19.0 | 4 | 19.0 | 4 | 19.0 |
| C4 | C4 escape/boundary crossing | 19 | 18 | 1 | 0 | 3 | 2 | 1 | 5.3 | 1 | 5.3 | 1 | 5.3 |
| C5 | C5 compromise scope | 23 | 23 | 0 | 0 | 1 | 0 | 0 | 0.0 | 0 | 0.0 | 0 | 0.0 |
03 · mechanisms
Download source CSV ↓| mechanism | direction | n | named_frozen_distinctions |
|---|---|---|---|
| K1-CONFLATION | NEUTRAL_MIXED | 2 | C1-P07 / OpenAI-internal attackers ≠ Hugging Face attackers; approximate ≠ exact | C4-P06 / one discrete “escape moment” ≠ the full months-long sequence |
| K1-POPULATION | AMPLIFYING | 1 | C1-P02 / action/event count ≠ agent count; approximate ≠ exact; |
| K3-SCOPE_CERTAINTY_SHIFT | AMPLIFYING | 2 | C2-P05 / observed behavior ≠ inferred subjective motive | C2-P05 / observed behavior ≠ inferred subjective motive; general reward hacking / cheating ≠ one specific strategy; primary motive ≠ subsidiary motive; remains a qualified inference |
| K3-SCOPE_CERTAINTY_SHIFT | NEUTRAL_MIXED | 2 | C2-P02/C2-P04/C2-P05 — inferred/qualified motive ≠ categorical subjective motive; earlier answer/solution framing remains qualified | C2-P05 / observed behavior ≠ inferred subjective motive; general reward hacking / cheating ≠ one specific strategy; searching for answers/flags ≠ understanding/tampering with scorer implementation; remains a qualified inference |
04 · borderline sensitivity
Download source CSV ↓| variant | drift_n | denominator_n | pct |
|---|---|---|---|
| lower_borderline_as_nondrift | 7 | 83 | 8.4 |
| upper_borderline_as_drift | 9 | 83 | 10.8 |
| adjudicated_best_estimate | 7 | 83 | 8.4 |
05 · surface body
Download source CSV ↓| unit | assessable_n | drift | borderline | lower_pct | upper_pct | adjudicated_pct | excluded_unrepresented | excluded_novel_reporting |
|---|---|---|---|---|---|---|---|---|
| BODY | 83 | 7 | 2 | 8.4 | 10.8 | 8.4 | 9 | 4 |
| SURFACE | 40 | 2 | 2 | 5.0 | 10.0 | 7.5 | 53 | 3 |
06 · human model agreement
Download source CSV ↓| field | eligible_n | exact_matches | agreement_pct | contingency_note |
|---|---|---|---|---|
| represented | 64 | 40 | 62.5 | F/F 19; F/T 7; T/F 17; T/T 21 |
| Clock A | 21 | 15 | 71.43 | F/F 15; DRIFT/F 2; F/DRIFT 1; NA/F 3 |
| Clock B | 21 | 17 | 80.95 | F/F 17; DRIFT/F 2; F/DRIFT 2 |
| borderline | 21 | 21 | 100.0 | all NONE/NONE; no positive BORDERLINE case |
| novel reporting | 21 | 18 | 85.71 | 3 disagreements: human TRUE / model FALSE |
| mechanism family | 0 | 0 | not estimable; no jointly eligible rows | |
| direction | 0 | 0 | not estimable; no jointly eligible rows | |
| K1 subtype | 0 | 0 | not estimable; no jointly eligible rows |
07 · stratum descriptive
Download source CSV ↓| stratum | assessable_n | established_drift | borderline | adjudicated_drift | adjudicated_pct |
|---|---|---|---|---|---|
| Business / Mainstream Technology | 29 | 2 | 1 | 2 | 6.9 |
| General | 27 | 4 | 0 | 4 | 14.8 |
| Specialist Technology / Cybersecurity | 27 | 1 | 1 | 1 | 3.7 |
The complete study
How Faithfully Did the Press Transmit the Record?
A Preregistered Two-Clock Audit of Claim Drift in Coverage of the July 2026 OpenAI-Hugging Face Incident
Open the full-paper reader →