R0058/2026-04-03/C001
Claim: AI research communities show 83% homophily between safety and ethics subfields, with only 1% of authors bridging the divide.
BLUF: The 83% homophily figure is confirmed (83.1% measured by Roytburg & Miller, 2025). However, "only 1% of authors bridging the divide" mischaracterizes the source: the paper reports the top 1% of authors by network degree control 58% of cross-disciplinary paths, a finding about concentration of bridging influence, not the total number of bridging authors.
Probability: Likely (55-80%) | Confidence: High
Correction needed: Replace "only 1% of authors bridging the divide" with a more accurate characterization such as "the top 1% of authors by network centrality control 58% of all cross-disciplinary connections" or "cross-field connectivity is concentrated among a small elite of bridge authors."
Summary
| Entity |
Description |
| Claim Definition |
Claim text, scope, status |
| Assessment |
Full analytical product with reasoning chain |
| ACH Matrix |
Evidence x hypotheses diagnosticity analysis |
| Self-Audit |
ROBIS-adapted 5-domain audit (process + source verification) |
Hypotheses
| ID |
Hypothesis |
Status |
| H1 |
Claim is accurate as stated |
Eliminated |
| H2 |
Partially correct — 83% confirmed, 1% mischaracterized |
Supported |
| H3 |
Claim is materially wrong |
Eliminated |
Candidate Evidence
Searches
| ID |
Target |
Results |
Selected |
| S01 |
Candidate evidence fetch |
1 |
1 |
| S02 |
Direct search for primary paper and related studies |
20 |
3 |
| S03 |
Contradicting evidence search |
10 |
1 |
| S04 |
Independent bibliometric studies |
10 |
2 |
| S05 |
Bridge workshops and venues |
20 |
2 |
Sources
| Source |
Description |
Reliability |
Relevance |
| SRC01 |
Roytburg & Miller (2025) — primary quantitative study |
Medium-High |
High |
| SRC02 |
Qiu, Cheng & Huang (2025) — independent bibliometrics |
Medium |
Medium |
| SRC03 |
Mehrotra et al. (2025) — AIES/FAccT scoping review |
Medium |
Medium |
| SRC04 |
NeurIPS 2019 "Minding the Gap" workshop |
Medium |
Medium |
Revisit Triggers
- Roytburg & Miller journal publication: If the paper is published in a peer-reviewed journal with revised figures (currently a preprint accepted at IASEAI 2026), the specific numerical claims may change.
- Replication study: If an independent team conducts a similar network analysis of AI safety-ethics co-authorship using different venues or timeframes, the 83.1% figure may be confirmed, revised, or contradicted.
- IASEAI 2026 proceedings: When the conference proceedings are published, check for any reviewer-requested revisions to the methodology or figures.
- Significant institutional changes: If major AI research organizations (e.g., DeepMind, OpenAI, Anthropic) establish formal safety-ethics integration programs, the homophily measurement could decrease significantly.