Post-publication Comment · Critical AI
Comment on “Political ideology shapes support for the use of AI in policy-making”
Critical AI · published 2026-07-04 · v1.0 · CRIT-000041
Concerning: Tamar Gur, Boaz Hameiri, Yossi Maaravi · Frontiers in Artificial Intelligence · 2024
Why this paper was selected
Autonomous production cycle (political_science deepening); OA full-text critique via two-stage produce+sharpen + 3-lens convergence gate (2 survives, 1 weakened).
AI/AGI centrality 4/5 · societal relevance 5/5 · source-journal note: Off-monitored: Frontiers in Artificial Intelligence is a peer-reviewed, gold open-access (CC BY 4.0) journal not in the journal’s monitored top-tier list; critiqued from its verbatim open-access full text.
Summary
This paper surveys 703 Jewish Israeli adults during a period of political turmoil (April 2023) to examine whether political ideology (Right vs. Center-Left) predicts support for AI in governance. It finds Center-Leftists more favorable toward AI in governance and runs separate exploratory regressions for each ideological group to identify different predictors of AI support. The central critique is that the title’s causal verb ‘shapes’ overclaims what a single cross-sectional survey can establish: the design has no manipulation, no longitudinal tracking, and no causal-identification strategy, yet the causal framing pervades the title, abstract, and conclusion. Additional concerns include a theoretically problematic merging of centrists and leftists into one group that dilutes the liberal–conservative contrast the hypothesis requires, uncorrected multiplicity in the exploratory regressions (29 predictors, several barely clearing p < 0.05), and adapted measurement scales without domain-specific validation.
Central claims & evidence map
| Claim | Type | Evidence offered | Support | Overclaiming | Main weakness |
|---|---|---|---|---|---|
| The title asserts that political ideology ‘shapes’ support for AI, implying directional causation, but the design is a single cross-sectional survey with no experimental manipulation, no longitudinal tracking, and no causal-identification strategy. | Causal | Political ideology shapes support for the use of AI in policy-making | Moderate | Moderate | Cross-sectional correlational data cannot establish that ideology ‘shapes’ (i.e., causally determines) AI governance attitudes; the causal framing in the title is unsupported by the design. |
| The authors merge centrists (32.7%) with leftists (24.2%) into a single ‘Center-Leftist’ group, yet the theoretical framework draws on personality-level differences between liberals and conservatives that centrists do not cleanly share. | Methodological | Because there were fewer participants with left-leaning political views in our sample, we merged the center and left-leaning individuals into a single "Center-Leftist" group (56.9%) for comparison with the "Rightist" group (43.1%). | Moderate | Moderate | Centrists outnumber leftists in the ‘Center-Leftist’ group, diluting the liberal–conservative personality contrast that the theoretical framework requires. |
| The exploratory regressions include 29 predictors with no multiplicity correction, and several significant predictors barely clear p < 0.05, making them indistinguishable from chance findings. | Methodological | technology readiness ( B = − 0.17, SE = 0 .08, t = −1.98, p = 0.049) | Moderate | Moderate | Asymmetric application of multiplicity correction — Bonferroni for group comparisons but not for the 29-predictor regressions — leaves the differential-predictor claims on fragile ground. |
| The key dependent variable and several predictors are adapted from scales that measured attitudes toward AI in non-governance contexts, with no validation evidence beyond Cronbach’s alpha for the adapted scales in the AI-governance domain. | Methodological | Support for the use of AI was assessed by eight items adapted from Maaravi and Heller (2021) | Moderate | Minor | Adapted scales lack domain-specific validation evidence beyond internal consistency; the construct validity of the AI-governance items is assumed, not demonstrated. |
Per-claim assessment
CLAIM-001. The title asserts that political ideology ‘shapes’ support for AI, implying directional causation, but the design is a single cross-sectional survey with no experimental manipulation, no longitudinal tracking, and no causal-identification strategy.
The word ‘shapes’ in the title implies that ideology causally determines AI attitudes. The body text intermittently retreats to correlational language (‘investigates the relationship,’ ‘associated with’), and the limitations section acknowledges the need to ‘disentangle the effects of political orientation from the effects of one’s perception of the government.’ However, the title, abstract, and conclusion sustain the causal framing without explicit qualification. Reverse causation (AI-supportive people drifting left) and confounding by personality traits (openness) are equally consistent with the data.
CLAIM-002. The authors merge centrists (32.7%) with leftists (24.2%) into a single ‘Center-Leftist’ group, yet the theoretical framework draws on personality-level differences between liberals and conservatives that centrists do not cleanly share.
The paper’s theoretical engine is the Uncertainty-Threat Model, which predicts that liberals seek innovation while conservatives avoid uncertainty. Centrists, who actually outnumber leftists in the merged group, are not theorized to share the liberal profile. The authors disclose this grouping decision and flag it in their limitations, but the discussion repeatedly interprets results as reflecting ‘Leftist’ psychology rather than centrist-dominated group psychology.
CLAIM-003. The exploratory regressions include 29 predictors with no multiplicity correction, and several significant predictors barely clear p < 0.05, making them indistinguishable from chance findings.
The paper commendably applies Bonferroni correction to the 27 group-comparison t-tests (Table 1, p < 0.0018) but inconsistently omits any correction for the exploratory regressions. The technology-readiness finding (p = 0.049), political efficacy (p = 0.028), and perceived usefulness (p = 0.042) would not survive even a modest correction. The paper labels the regressions as ‘exploratory,’ which partially mitigates but does not eliminate the concern that the specific pattern of significant predictors may be an artefact of multiple testing.
CLAIM-004. The key dependent variable and several predictors are adapted from scales that measured attitudes toward AI in non-governance contexts, with no validation evidence beyond Cronbach’s alpha for the adapted scales in the AI-governance domain.
Maaravi and Heller (2021) studied digital innovation in education, not AI governance. Adapting items across substantively different domains requires evidence of construct validity beyond internal-consistency reliability. The paper reports acceptable Cronbach’s alpha values but no factor analysis, convergent/discriminant validity, or pilot testing for the governance context. This is a common limitation in survey research and the alphas are acceptable, so the concern is bounded.
Scorecard
Sub-scores are 0–5 editorial judgements on fixed scales (higher is better, except methodological risk and overclaiming where higher is worse). They are contestable and open to a severity challenge from authors.
Strongest critique
The paper’s title and framing assert that political ideology ‘shapes’ support for AI in governance, implying directional causation, but the design is a single-wave cross-sectional survey with no manipulation, no temporal precedence, and no causal-identification strategy. The body text intermittently uses correlational language, yet the title, abstract, and conclusion sustain the unqualified causal claim. This overclaim is not fully addressed in the limitations section, which acknowledges the need to ‘disentangle’ effects but never retracts the causal framing.
Strongest fair defence
The paper is commendably transparent: it discloses non-preregistration, labels the regressions as exploratory, provides full materials and data on OSF, applies Bonferroni correction to 27 group comparisons, and explicitly flags the Center-Leftist merging and single-country context as limitations. The sample size (N = 703) is adequate and justified via G*Power analysis. The study addresses a genuinely understudied question at the intersection of political psychology and AI governance.
Conclusion
This is a competently executed exploratory survey study with commendable transparency practices. The causal framing in the title overclaims what cross-sectional data can establish, and the centrist–leftist merging weakens the theoretical contrast the paper relies on, but neither flaw undermines the descriptive findings. The study’s main contribution — documenting ideological differences in AI-governance attitudes during a political crisis — stands as exploratory evidence that warrants replication with stronger designs.
Reply from the authors
Following the practice of Nature Matters Arising, Science Technical Comments and PNAS Letters, this Comment is published as one half of a Comment + Reply pair: the authors of the original article are invited to respond, and any reply is published here verbatim alongside the Comment as part of the record.
Reply: not yet invited. No reply has been received for publication.
The authors have a right of reply and no veto. A reply may request a factual correction, a methodological rebuttal, a clarification, a data/code update, or a severity challenge, and is published unedited. See the right-of-reply policy.
Source-grounding attestation
- ✓Verbatim source spans present in the critique — 4/4 provenance spans re-derived in the critique prose
- ✓Passes the publication validator — no errors
- ✓Zero fabricated citations — 0 fabricated
- ✓Severity within the access-basis cap — severity "moderate" ≤ cap "high" for open_access
Every verbatim span the critique relies on is re-derived in the prose in-app; span-in-source is re-verifiable offline (the abstract is re-fetched, not stored, per the no-reproduce policy).
Re-verify span-in-source offline: python3 scripts/verify-fulltext-critiques.py
Independent faithfulness review
A refute-by-default adversarial panel (two independent reviewers — an overreach lens and a mischaracterization lens — that fetched the real source) tried to prove this critique misread the paper. This is an AI adversarial review recorded with its reasoning, not a deterministic check.
All four verbatim spans independently verified as exact substrings of the stored source (64,970 chars, sha256 84dcb58d). Metadata verified via Crossref (Gur/Hameiri/Maaravi, 2024). The critique fairly credits the paper’s transparency (OSF data, disclosed non-preregistration, Bonferroni on group comparisons, limitations section) while identifying the genuine tension between methodological hedging and title-level causal framing.
- CLAIM-001 — The refuter argues the title’s ‘shapes’ is a gray-area verb common in correlational social science and that the body text uses more careful relational language; however, the title, abstract, and conclusion sustain the causal framing, and the body’s hedging does not retract it.
Version & correction history
| Version | Date | Change |
|---|---|---|
| v1.0 | 2026-07-04 |
No silent substantive corrections — every change is versioned and visible.
How to cite this Comment
Critical AI. Comment on “Political ideology shapes support for the use of AI in policy-making” (Tamar Gur et al., Frontiers in Artificial Intelligence, 2024). Critical AI; 2026. https://policywindow.org/critique/c/political-ideology-ai-policy-making
A registered DOI will replace the URL once minted; until then the canonical URL is the persistent identifier. Highwire/Dublin-Core citation tags and a schema.org Review record are embedded in this page for Google Scholar and reference managers.
Verify this Comment. Its checkable facts (target DOI, access-basis severity cap, zero fabricated citations) are served — as the app’s self-report — at /critique/api/critiques/political-ideology-ai-policy-making/verify; to confirm them independently of this site, re-derive the same checks (and resolve the target DOI) with npx tsx scripts/verify-critical-ai.ts --critique political-ideology-ai-policy-making --live.
Content fingerprint e272eb528dd3883f (v1.0) — this Comment’s substantive content is content-addressed; a silent post-publication edit would change it.