The three 2026 warp papers are one correlated source: same single author, arXiv preprints without journal DOIs, zero citations each on OpenAlex (reproduced live with a positive control: S-0001 shows 45). Within that caveat, S-0003 certifies the Van Den Broeck walls as Hawking-Ellis Type IV — no rest frame — ABOVE a Type-I to Type-IV transition at v_s ~ 0.36 (the value lives in a source macro; the canonical text loses it), with Alcubierre and Natario walls Type-IV dominated at all speeds; and S-0005's 'no exotic matter' drive is, in its own words, causal, subluminal, and steeply but positively costly.
What it rests on
Attacks run against it
Ran all four vectors. (A2, survived) Reproduced the citation prong live on OpenAlex 2026-08-10: title.search for each of the three 2026 titles returned only records with cited_by_count=0, sole author An T. Le / A. T. Le, and no journal DOI (only 10.48550/arxiv.* DataCite and Zenodo DOIs); positive control https://api.openalex.org/works/https://doi.org/10.1103/PhysRevD.105.064038 returned cited_by_count=45, exactly the claim's pinned value. Verified single authorship directly in all three corpus main.tex author blocks (identical name, affiliations, emails). Verified the macro parenthetical: corpus/S-0003/main.tex line 29 is \newcommand{\vdbTransitionVS}{0.36}, and canonical fulltext.md contains zero occurrences of "0.36" and two unexpanded "vdbTransitionVS" tokens. (A3, decisive) Compared the claim's Van den Broeck conjunct against E-0031's span and its surrounding canonical fulltext: the source sentence, by ellipsis, says the VdB walls are "Type-IV DOMINATED" above the transition, and at fulltext char ~36957 defines the transition as "its wall Type I fraction crossing 50 [%] at v_s≈vdbTransitionVS". The claim instead asserts the walls are "Hawking-Ellis Type IV — no rest frame — ABOVE" the transition, and by inverting the clause order (VdB first, Alc/Nat with "Type-IV dominated" second) it severs the ellipsis and reads as a deliberate contrast: VdB fully Type IV vs Alc/Nat merely dominated. The source's own certified numbers falsify that reading: Table tab:velocity_type gives VdB Type I = 12.2% (v_s=0.5), 18.3% (1.0), 25.6% (2.0), and the text says the Type-I fraction "rises monotonically, reaching ... 29.6% (Van den Broeck) at v_s=2.5" because "more of the wall acquires a rest frame as the bubble speeds up" — so above 0.36 a substantial and growing share of the wall HAS a rest frame. Reproduction script: attack_c0022.py in the scratch dir (run from the repo root). This is the same failure class (dropped source qualifier) for which X-0023 refuted this claim's direct parent C-0016. || FINDINGS: Refuting facts: (1) canonical fulltext S-0003 @ ~36957: "Van den Broeck additionally passes through a velocity-driven Type-I to Type-IV transition, its wall Type I fraction crossing 50 at v_s≈vdbTransitionVS" — the transition is a 50%-majority crossing, not an onset of all-Type-IV; (2) body.tex Table tab:velocity_type: VdB Type I/IV = 12.2/87.8 (v_s=0.5), 18.3/81.7 (1.0), 25.6/74.4 (2.0); (3) "The Type-IV fraction decreases with v_s (more of the wall acquires a rest frame as the bubble speeds up)" — the claim's "no rest frame" appositive is false of 12-30% of the wall above the transition, and the trend runs opposite to the claim's implication. Surviving prongs: single author An T. Le confirmed in all three main.tex files; OpenAlex live 2026-08-10 gives cited_by_count 0/0/0 with no journal DOIs (only arXiv/Zenodo DOIs) and positive control 10.1103/PhysRevD.105.064038 = 45, matching Q-0002; main.tex:29 = \newcommand{\vdbTransitionVS}{0.36} with "0.36" absent from canonical fulltext (2 unexpanded macro tokens); Alc/Nat "Type-IV dominated at all speeds" holds (Natário worst case 61.3% Type IV at v_s=2.5); S-0005 quotes E-0026/E-0027 are exact ("causal, subluminal... no exotic matter in the thin-shell idealization"; "cost is steep, but positive and finite"). Fix for the superseding claim: render VdB as "Type-IV dominated above a Type-I→Type-IV transition (wall Type-I fraction crossing 50% at v_s≈0.36)", attach "no rest frame" to Type-IV points, and optionally carry the decreasing-fraction direction. || ADJUDICATOR (independent second opinion, verdict UPHELD): Attacker's REFUTED verdict stands. I re-ran the decisive A3 step directly against the repo (corpus/S-0003) and via the attacker's script (/tmp/.../redteam34/C-0022/attack_c0022.py); every fact reproduces. (1) The source predicate for Van den Broeck is "Type-IV dominated", not "Type IV": the abstract sentence containing E-0031 (fulltext ~char 4553) elides "are Type-IV