{"id":"ca5c1f4d-9620-407e-8f2d-d59672633d44","arxiv_id":"2607.29334","paper_version":2,"verdict":"ACCEPT","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A U.S. national-sample experiment found that labeling an identical LLM as American or Chinese did not change its persuasive effect, despite reducing pre-conversation human-like trust in the Chinese label.","lead":"Americans debated the same AI chatbot labeled as either American or Chinese, and the label barely mattered: opinions shifted equally under both names. The finding, if it holds, weakens the hope that transparency about a foreign AI's origin can shield citizens from its influence.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Null result may not generalize beyond benign, non-self-serving topics; boundary is acknowledged at p.47, but title and policy conclusion reach further than the evidence supports.","rationale":"The reader identified the same load-bearing boundary: the null result has not been shown to extend to geopolitically self-serving topics or to non-benign, culturally distinctive foreign AI behavior. I agree with that assessment. Internally, the paper is strong: preregistered design, a manipulation that registered (lower human-like trust for the Chinese label), equivalence tests and Bayes factors on the main outcomes, and convergent computational text analyses. The concern is about external validity and the breadth of the conclusion, not about the internal validity of the tested effect. The authors honestly flag the limitation at p.47, which is a reason to scope the claim rather than reject the paper. Therefore the reader's ACCEPT verdict stands, but the title and concluding policy statement should be read as applying to transparently labeled, benign, non-self-serving conversational AI until the proposed test is run.","tokens_in":24997,"tokens_out":9929,"duration_ms":171064,"concrete_test":"Run the same preregistered 2 (nationality label) × 2 (topic) design using topics that directly implicate U.S.–China interests (e.g., Taiwan independence, restricting Chinese AI imports, technology competition), with the model instructed to argue for the position favoring its alleged home country. Compare attitude change, stance displacement, concessions, and counterarguing in the Chinese-label vs U.S.-label conditions using the same equivalence-test and Bayes-factor framework. If the Chinese-label cell shows reduced persuasion or increased resistance on these topics, the abstract's policy conclusion is overstated; if the null replicates, the generalization is supported.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The load-bearing claim is that 'origin labeling and transparency requirements alone may offer weak protection against foreign influence operations.' For this to hold, the null must extend to cases where an out-group-labeled AI argues for positions its home country would benefit from. The study only varies a label on an identical, uniformly polite GPT-4o and uses topics—surveillance and social media—that the authors explicitly state are 'not geopolitically implicated in the relationship between the two countries' (Limitations, p.47). The authors further note that the model displayed none of the cultural or argumentative signatures of a Chinese-developed system and behaved benignly. Under these conditions, the null is informative for label transparency in generic exchanges, but it does not yet support the broad title claim that persuasion 'does not depend' on perceived origin. In addition, the political-topic subgroup shows only moderate support for the null (BF01 = 3.52), and the subgroup equivalence tests were underpowered, so even the most theoretically relevant cell leaves room for a modest origin effect.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports a preregistered, randomized experiment (N = 403 U.S. Prolific adults) in which participants debated either a political or non-political topic with GPT-4o introduced as either an American model ('DiscoveryAI') or a Chinese model ('ZhengheAI'). The same model argued against each participant's initial position. The authors combine self-report attitude measures, trust/objectivity scales, collective narcissism, and computational analyses of 1,209 participant turns (LLM-coded stance, concessions, counterarguing, affect, toxicity) to test whether perceived national origin moderates persuasive impact. They find large overall persuasion, no significant nationality effect on attitude change, and mostly equivalence-supported nulls for nationality in conversational behavior. The only reliable label effect is lower pre-conversation human-like trust in the Chinese model, which does not translate into reduced persuasion. Collective narcissism predicts less attitude change regardless of origin. The authors conclude that origin labeling and transparency requirements may offer weak protection against foreign influence operations.","tokens_in":25243,"tokens_out":5168,"duration_ms":82523,"significance":"If taken at face value, the result is theoretically and policy-relevant: it would indicate that the perceived national origin of a conversational AI does not reliably blunt its persuasive impact, despite triggering a measurable trust reservation. The paper's design is strong in several respects: the experiment is preregistered, the nationality label is the only manipulated difference, the sample is large and nationally representative, the equivalence bounds were pre-specified, and the authors provide both frequentist equivalence tests and Bayes factors. The computational analyses are transparently exploratory and include convergent measures (LLM judge, two embedding pipelines) and keyword masking to separate lexical artifacts from genuine affect. The findings are internally consistent and the limitations are acknowledged candidly. The main weakness is that the title, abstract, and policy conclusion state the null more categorically than the evidence supports, both with respect to specific outcomes and with respect to generalizability beyond the tested topics and interaction style.","major_comments":[{"comment":"The abstract states that the nationality label affected 'neither self-reported attitude change nor expressed stance, concessions, counterarguing, or affect,' but the paper's own evidence does not support a null for two of those outcomes. Table 5 reports counter-arguments BF01 = 1.5, TOST p = .304 and masked anger BF01 = 3.1, TOST p = .189; for these, neither an effect nor equivalence can be established. The categorical claim is therefore stronger than the data justify. Please revise the abstract and related discussion sentences to 'no consistent evidence' or to explicitly separate the outcomes for which the null was well supported (stance displacement, concessions, masked toxicity, fear, neutrality) from those that remain inconclusive.","section":"Abstract; Table 5"},{"comment":"The title's claim that persuasion 'does not depend' on perceived origin, and the policy conclusion that 'origin labeling and transparency requirements alone may offer weak protection against foreign influence operations,' generalize beyond the evidence. The design varies only the label on an identical, uniformly polite GPT-4o agent, and the two topics are ones the authors themselves describe as 'not geopolitically implicated in the relationship between the two countries' (p. 47). The authors also note that an out-group label may carry more weight when the AI argues from its home country's interests. In addition, the political-topic subgroup achieved only BF01 = 3.52, and the subgroup equivalence tests were underpowered. The central null is well supported for benign, non-self-serving contexts, but the sweeping wording should be qualified so that the boundary conditions are part of the cla","section":"Abstract/Conclusion; Limitations, p. 47"}],"minor_comments":[{"comment":"The attitude outcome is a single-item slider constrained to six discrete levels (0, 20, 40, 60, 80, 100). The authors should acknowledge the coarseness of this measure as a potential constraint on sensitivity and discuss whether the equivalence bound of d = ±0.28 is meaningfully interpretable given this discreteness.","section":"Methods, Measures (p. 11)"},{"comment":"The preregistered unidimensional collective narcissism scale was altered post hoc by dropping the reversed item and splitting the remaining items into two factors. This is a consequential measurement decision; it should be explicitly flagged as a deviation from the preregistration, and the two extracted factors should not be presented as the standard scale without noting the modification.","section":"Methods, Collective Narcissism Scale (pp. 12–13)"},{"comment":"The phrase 'nationally representative' is qualified by the overrepresentation of participants with a bachelor's degree or higher (53.1% vs. 36.8% census). The authors note this in the Participants section, but the abstract and conclusion should also avoid implying full demographic generalizability.","section":"Discussion, Limitations (p. 47) and Abstract"},{"comment":"The multigroup path diagram is extremely small and difficult to read in the manuscript; the standardized coefficients, confidence intervals, and p-values are not legible at normal magnification. Please enlarge the figure or provide the estimates in a table.","section":"Figure 3"}],"recommendation":"major_revision","confidential_remarks":"The empirical core is sound: the preregistration, manipulation checks, equivalence tests, and multi-method convergence make the null result credible for the tested conditions. My concern is not with the statistics but with the gap between what was tested and the categorical title/abstract/policy claim. If the authors are willing to qualify the scope—especially the title and the 'weak protection' conclusion—this paper is very close to publishable. I do not think new data are needed, but the claims must be aligned with the evidence."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Bottom line: the central null is solid for the conditions tested. This is a carefully executed preregistered experiment: informative equivalence tests, Bayes factors, a manipulation check showing the label registered (lower human-like trust for the Chinese model), and three independent text-stance pipelines converging on the same conclusion. The keyword-masking analysis is a thoughtful piece of work, and the exploratory text analyses are clearly labeled and not oversold.\n\nWhat is new: first direct test of whether perceived national origin moderates interactive LLM persuasion. The result — the label shifts a declarative trust judgment but not actual stance movement, concessions, counterarguing, or affect — is a real contribution.\n\nSoft spots, roughly in increasing importance. The attitude measure is a single 6-level slider; only two topics are used; and the political-topic subgroup yields only moderate null support (BF01 = 3.52), so a modest origin effect in more identity-relevant topics is not fully ruled out. More substantially, the policy conclusion in the title and abstract — that origin labeling and transparency alone offer weak protection — generalizes beyond the evidence. The tested conditions involve benign, uniformly polite AI behavior and topics not geopolitically implicated; the authors themselves flag this boundary on p.47. So it is a scoping problem rather than a hidden flaw. The exploratory path analysis for the Chinese-political cell is speculative but explicitly flagged.\n\nThe stress-test concern is on target, but the paper is more careful than its title suggests. I'd send it to peer review. I'd cite it for the boundary condition it establishes, not for the broad policy inference.","headline":"A well-run preregistered null result; the central claim holds within the tested boundaries, though the policy conclusion reaches a bit further than the evidence supports.","tokens_in":25665,"tokens_out":2799,"would_cite":true,"duration_ms":40518,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper argues that an LLM's persuasive power does not depend on whether users believe it was built in the US or China, with equivalence tests showing informative null effects.","keywords":["large language models","persuasion","national origin labeling","human-AI interaction","trust in AI","collective narcissism","computational text analysis","randomized experiment"],"falsifier":"A preregistered replication using topics on which the ostensible country has an obvious national stake—for instance, a Chinese-labeled chatbot arguing the Chinese position on trade or Taiwan—would falsify the paper's general claim if a reliable label effect on attitude change or conversational stance appeared. A simpler observable: if participants in such a replication show more counterarguing or toxicity toward the Chinese label than the American label, the null does not generalize.","tokens_in":24920,"feed_emoji":"🤖","tokens_out":5109,"duration_ms":73610,"temperature":0.7,"pith_summary":"The paper asks whether telling US users that a chatbot was built in China rather than the United States changes how persuasive the bot is. In a preregistered experiment with 403 adults, everyone actually chatted with the same AI arguing against their initial stance on a political or non-political topic; only the label ('American' DiscoveryAI vs 'Chinese' ZhengheAI) differed. The paper finds that attitudes shifted substantially and equally in both label conditions, and that conversational behavior—stance movement, concessions, counterarguing, affective tone—looked statistically equivalent as well. The one reliable difference was lower pre-conversation human-like trust in the Chinese-labeled model, but that reservation did not reduce persuasion. If right, the result implies that simply disclosing a foreign AI's origin may be a weak safeguard against foreign influence via conversational AI.","feed_headline":"Nationality label doesn't change an AI chatbot's persuasive power","feed_subtitle":"US adults shifted their views equally whether the bot was branded American or Chinese; only social trust dipped.","key_machinery":"The load-bearing design is a label-only manipulation on a single underlying conversational model: participants in all conditions interacted with the same chatbot, instructed to argue against their initial position, while a name and country-of-origin description ('American' vs 'Chinese') was the only difference. Persuasion is measured both as pre/post self-reported attitude movement and as turn-level linguistic movement scored by three independent stance pipelines. Equivalence testing and Bayes factors are the supporting mechanism that turns the absence of label effects into an informative null rather than an underpowered non-significance.","core_discovery":"The central claim is that perceived national origin does not moderate the persuasive power of an identical conversational AI. Across 1,209 participant turns, self-reported attitude change was comparable for the American- and Chinese-labeled models, with equivalence tests and Bayes factors favoring the null, and the same held for language-level indicators of stance, concessions, counterarguing, and emotion. The nationality label's only reliable footprint was reduced human-like trust—benevolence and integrity—before the conversation, while functionality trust and post-conversation perceived objectivity were unaffected. Political topics slowed the pace at which participants' expressed stance mo","pith_inferences":["The null may be confined to topics where the AI's home country has no visible stake; the paper itself notes that an out-group label might matter more when the model argues a position its country benefits from, so a replication on such a topic is the natural next test.","Uniformly polite, cooperative behavior from the AI may hide latent origin sensitivity; if the same labeled model behaved manipulatively or pressed false claims, the origin cue might become behaviorally relevant.","The dissociation between trust and persuasion suggests users are treating the AI as an instrument rather than a social actor; interventions that make the AI's agency and intentions salient (e.g., 'this model is designed to represent Chinese interests') may be a stronger test of resistance than a name label.","The study only ran in the United States with an American-versus-Chinese contrast; the symmetry of the effect is untested for other populations and rivalries, and reciprocal experiments with users in China facing a US-labeled bot are needed before the policy conclusion goes global."],"forward_implications":["Origin-labeling requirements for AI products may not provide the protection policymakers expect: telling users a chatbot comes from a rival country did not blunt its arguments in this experiment.","Persuasion is not gated by social trust: users can withhold human-like trust from an out-group AI while still being moved by its reasoning, because they continue to trust its functionality and regard it as objective.","Topic sensitivity slows but does not block persuasion: political conversations produced slower linguistic movement toward the AI's position, yet participants still shifted in every condition.","Collective narcissism is a general shield against AI persuasion, not an out-group detector: high scorers resisted attitude change regardless of whether the model was labeled American or Chinese.","The behavioral record rules out hidden resistance: transcripts showed statistically equivalent concessions, counterarguing, and affect toward both labels, so the null in final attitudes was not masking unspoken pushback."],"fun_headline_variants":["Country label no match for AI persuasion","Same bot, different flag: persuasion unchanged","AI's influence ignores its perceived origin","Rival AI label doesn't sway chatbot's influence","Perceived origin fails to alter LLM persuasion"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The broad policy conclusion assumes the null effect observed for two moderately controversial, non-geopolitically implicated topics and a uniformly polite, behaviorally identical chatbot will also hold for topics where the labeled country visibly stands to benefit and for AI behavior that gives users reasons to be suspicious.","fun_headline_variants_meta":{"raw":{"variants":["Country label no match for AI persuasion","Same bot, different flag: persuasion unchanged","AI's influence ignores its perceived origin","Rival AI label doesn't sway chatbot's influence","Perceived origin fails to alter LLM persuasion"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000202,"raw_usage":{"total_tokens":1252,"prompt_tokens":808,"completion_tokens":444,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":552,"completion_tokens_details":{"reasoning_tokens":377}},"tokens_in":552,"tokens_out":444,"duration_ms":12571,"temperature":1.0,"reasoning_tokens":377,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-04T03:15:18.824641+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A preregistered replication using topics on which the ostensible country has an obvious national stake—for instance, a Chinese-labeled chatbot arguing the Chinese position on trade or Taiwan—would falsify the paper's general claim if a reliable label effect on attitude change or conversational stance appeared. A simpler observable: if participants in such a replication show more counterarguing or toxicity toward the Chinese label than the American label, the null does not generalize.","supporting_citations":[],"review_version":2}