{"id":"9d294f6d-3bb3-4c0f-9808-6b318b6c3143","arxiv_id":"2506.09185","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"An autoethnographic study of 21 AI education stakeholders concludes that whole-person education is essential for preparing ethical and socially responsible AI engineers.","lead":"This paper argues that AI engineering education should incorporate whole-person education, integrating ethics and interdisciplinary perspectives into technical training. It bases this argument on the collective autoethnographic reflections of the 21 authors, who are also the study participants.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The study's design guarantees its conclusion: author-participants are self-selected whole-person advocates, so their collective reflections cannot independently establish a 'pressing need' for whole-person AI education.","rationale":"The reader correctly identifies the self-selection of the 21 author-participants as the weakest assumption behind the general claim. My concern goes one step further: the study's research questions and prompts are explicitly framed around whole-person education advocates, and the Appendix shows unanimous endorsement. This is not merely a representativeness problem; it is a circularity problem, because the 'findings' are the output of a selection process that presupposes the conclusion. The paper can legitimately describe what motivates this particular group of advocates and how they envision the future, and the reader's CONDITIONAL verdict already demands tempering of generalizability claims. My concern reinforces that demand rather than moving the verdict. I would keep the verdict at CONDITIONAL, with the explicit condition that the abstract and discussion be rewritten to scope all findings to self-selected advocates, and that the participant-count inconsistencies be resolved. No external empirical evidence is required for the paper to stand as an advocacy piece, but the central claim as currently worded overstates the epistemic status of the reflections.","tokens_in":26996,"tokens_out":4301,"duration_ms":46576,"concrete_test":"Conduct a systematic text audit of the Appendix and Fig. 3: extract each reflective prompt and each author-participant's stance on whole-person education, coding for endorsement, skepticism, or conditionality. Then check whether any relationship exists between the prompts' phrasing and the uniformity of endorsement. If every prompt presupposes advocacy and every response endorses it, the central claim is an artifact of selection and should be reframed as 'self-selected advocates call for...' rather than 'findings strongly reaffirm a pressing need.' As a second check, reconcile the participant count across the abstract (14), Section III (21, then 20), and the Appendix (21); if the count cannot be reconciled, the method section cannot be verified.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central claim, stated at the opening of Section IV, is that 'our findings strongly reaffirm a pressing need for a holistic, interdisciplinary AI education.' For this claim to be supported by the study, the author-participants' reflections would need to constitute evidence of need. But the design precludes that: RQ1 explicitly targets 'whole-person education advocates,' and Section III describes the research team as self-selected participants who 'affirm that subjectivity is a feature' of the method. The Appendix contains 21 uniformly affirming narratives with no reported dissenting, skeptical, or ambivalent voice, and the reflective prompts (Fig. 3) ask participants to elaborate on their advocacy for whole-person education. The 'findings' are therefore restatements of the selection criterion, not independent evidence of a pressing need. The Limitations section cautions about generalizability, but the abstract and Section IV use unqualified language ('AI engineers are equipped not only with...', 'a pressing need'), overreaching what an autoethnography of advocates can establish. The result is a circular argument: advocates of whole-person education reaffirm the need for whole-person education. This does not invalidate the paper's value as a position statement, but it means the central claim cannot be taken as a research finding without external evidence such as curriculum audits, employer demand analyses, or student outcome data.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper reports a collaborative autoethnographic study of the need for whole-person education in AI engineering education. Twenty-one (elsewhere fourteen or twenty) author-participants from academia, industry, and non-engineering fields contributed written reflections and took part in two synchronous sessions. Using thematic analysis, the authors identify five motivations (global and culturally responsive education, bridging academia and industry, ethics as foundational, interdisciplinary learning, democratization of AI education) and five future-oriented visions (challenging technological neutrality, moving beyond technosaviourism, centering interdisciplinarity, user-centered curricula, ethical leadership). The paper concludes that the findings strongly reaffirm a pressing need for holistic, interdisciplinary AI education and offers implications and recommendations for curriculum transformation.","tokens_in":27338,"tokens_out":3636,"duration_ms":40231,"significance":"If the central claim were supported, the paper would make a useful contribution to engineering education discourse by foregrounding ethical, social, and interdisciplinary dimensions of AI training from a diverse set of global voices. The paper's strengths are its transparent use of collaborative autoethnography, its explicit affirmation of subjectivity as a feature of the method, its geographically diverse author-participant group, and its rich appendix of personal reflections. It also usefully challenges technological neutrality and technosaviourism. However, the evidence base is a self-selected group of whole-person education advocates reflecting on their own advocacy, which cannot independently establish a pressing need. The paper is best read as a collective position statement or illustrative qualitative inquiry; the generalizing language in the abstract and discussion overreaches the method's evidentiary scope.","major_comments":[{"comment":"The central claim that 'our findings strongly reaffirm a pressing need for a holistic, interdisciplinary AI education' is not supported by the study design. RQ1 explicitly targets 'whole-person education advocates,' the author-participants are self-selected advocates, Figure 3's prompts ask participants to elaborate on their advocacy, and the appendix contains no dissenting, skeptical, or ambivalent voices. The conclusion therefore largely restates the selection criterion rather than independently establishing a 'pressing need.' The Limitations section (Section VI) appropriately notes that the method does not aim for generalizability, but the abstract and Section IV use unqualified language ('AI engineers are equipped not only with...', 'a pressing need'). I recommend reframing the findings as the collective perspective of a self-selected group of advocates, or supplementing the study with external evidence such as curriculum audits, employer-demand analyses, or student outcome data before making the general claim.","section":null},{"comment":"The participant count is inconsistent: the Abstract says 'fourteen diverse stakeholders,' Section III says 'The research team consisted of twenty one participants,' and later the same section states that 'twenty participants initially responded' and that additional author-participants 'bring[ing] the total number of contributing participants to twenty.' Because the author-participants' reflections are the entire dataset, this discrepancy is not cosmetic; it obscures the evidentiary basis of the study. Please reconcile the participant count throughout and report exactly how many participants contributed at each stage.","section":null},{"comment":"The thematic analysis is described as following the six-phase process of Braun and Clarke [49], but the paper does not report how coding disagreements were resolved, how the synchronous sessions contributed to theme validation, or whether analytic memos or coding trails were produced. The claim of enhanced 'trustworthiness' through collective reflexivity is asserted but not evidenced. Given that the central finding rests entirely on the authors' own analysis of their own reflections, a brief account of the coding process, disagreement resolution, and reflexive checks would strengthen the manuscript.","section":null}],"minor_comments":[{"comment":"The phrase 'Engineers’s decisions' contains a typo and should be 'Engineers’ decisions.'","section":null},{"comment":"The phrase 'technosavvy' appears where 'technosaviourism' is clearly intended; please correct the terminology for consistency with Section II.","section":null},{"comment":"The citation placeholder '[cite: , preparatory, and secondary level' is incomplete and should be replaced with the actual reference or removed.","section":null},{"comment":"Reference [35] ('B. RADELJIC, AI as a new public intellectual?') lacks publication venue, year, and page or DOI details; please complete it.","section":null},{"comment":"The caption states the prompts 'highlighting their advocacy for whole person education,' which makes the confirmatory framing explicit; consider rewording the caption to neutrally describe the prompts themselves.","section":null}],"recommendation":"major_revision","confidential_remarks":"The core issue is the gap between the paper's stated autoethnographic method and its generalizing claims. If the authors reframe the paper as a collective position statement and reconcile the participant counts, it could become a valuable contribution to the CEEA-ACÉG conversation. I see no indication of authorship or citation misconduct; the author-participant structure is inherent to collaborative autoethnography, but its implications for the paper's claims need to be addressed honestly."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this is a readable, honest advocacy paper that makes no pretense of being a controlled study, but the abstract and Section IV claim more than the design can support. The authors are 21 self-selected whole-person education advocates whose autoethnographic reflections are the data; the findings restate the selection criterion. That is not a fatal flaw for a position piece, but it means the central claim is not a research finding.\n\nWhat's good: the paper is transparent about method, gives a genuinely diverse set of voices across geography and career stage, and the literature review on technological neutrality and technosaviourism is competent. The appendix is rich material for anyone designing ethics-and-AI curriculum discussions. The authors' own limitations section is appropriately cautious.\n\nSoft spots: (1) participant count says 14 in the abstract, 21 in the method, then 20 twice; (2) an unresolved citation placeholder in Appendix G ('[cite: , preparatory, and secondary level'); (3) the abstract's 'findings strongly reaffirm a pressing need' is not supported by an autoethnography of advocates; the method section itself says generalizability is not a goal. The stress-test note is essentially correct. The remedy is to reframe the paper as a position statement or 'collective reflection' rather than as evidence of need, and to correct the factual slips.\n\nWho it's for: engineering education researchers interested in autoethnographic methods or in arguments for integrating ethics into AI curricula. It could be a useful reading-group piece. As a research contribution, it's a call to action with illustrative narratives, not a demonstration.\n\nI'd send it to peer review if it's positioned as a perspective piece; as a research paper, it needs major revision first. Worth engaging.","headline":"Honest, diverse advocacy for whole-person AI education, but its central claim is a restatement of the authors' own selection, so it reads better as a position statement than as research findings.","tokens_in":27847,"tokens_out":1707,"would_cite":false,"duration_ms":18404,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Whole-person education—ethics, global perspective, and interdisciplinary collaboration—should be core to AI engineering curricula, argue twenty educators and practitioners through collaborative autoethnography.","keywords":["whole-person education","AI engineering education","collaborative autoethnography","technological neutrality","technosaviourism","ethics education","interdisciplinary curriculum"],"falsifier":"A comparative study tracking AI engineering graduates from whole-person-integrated programs versus traditional technical programs could test the claim: if both groups show equivalent ethical reasoning, interdisciplinary collaboration, and societal impact in their work, the paper's central premise would not hold. Alternatively, a large-scale survey of AI engineering educators and practitioners asking whether they see the same gaps in current curricula would provide evidence for or against the universality of the reported motivations.","tokens_in":26793,"feed_emoji":"🎓","tokens_out":4151,"duration_ms":37958,"temperature":0.7,"pith_summary":"This paper argues that current AI engineering education, by focusing almost exclusively on technical proficiency, produces engineers unprepared for the ethical and societal weight of their work. Drawing on collaborative autoethnography with twenty international author-participants, it claims that whole-person education—integrating ethics, global perspectives, interdisciplinary collaboration, and social responsibility—should be a core part of AI engineering curricula. The authors challenge the idea that technology is neutral and reject technosaviourism, arguing instead that AI systems embody the values and power structures of their creators. If the finding holds, AI engineering programs would need substantial reform in how they teach, evaluate, and accredit future engineers.","feed_headline":"AI engineers need whole-person training, not just code","feed_subtitle":"A 20-author autoethnography finds ethics, global perspective, and interdisciplinary skills belong at the core of AI curricula.","key_machinery":"The methodological machinery is collaborative autoethnography (CAE) combined with reflexive thematic analysis. Twenty author-participants from diverse global backgrounds, acting as both researchers and data sources, produced written reflections responding to prompts and then engaged in dialogic and synchronous sessions. The six-phase thematic analysis process (familiarization, coding, generating themes, reviewing, defining, writing up) organizes these narratives into the paper's motivational and visionary themes. Conceptually, the paper uses whole-person education as the theoretical lens, connecting it to value-sensitive design, inclusive design, and participatory design frameworks.","core_discovery":"The central claim is that whole-person education is necessary for AI engineering education. The paper states that \"our findings strongly reaffirm a pressing need for a holistic, interdisciplinary AI education.\" Participants' reflections, analyzed through thematic analysis, yield five motivations—global and culturally responsive education, bridging academia and industry, ethics as foundational, interdisciplinary learning, and democratization of AI education—and a vision for curricula that challenge technological neutrality, move beyond technosaviourism, center interdisciplinarity, design user-centered inclusive curricula, and prioritize ethical leadership and lifelong learning. The paper presents this as a call to reconceptualize engineering education so that AI engineers act as stewards of sociotechnical systems rather than neutral builders.","pith_inferences":["If whole-person education becomes standard, accreditation bodies and industry hiring criteria may need to define and assess competencies like ethical discernment and interdisciplinary teamwork, which are harder to measure than technical skills.","The same autoethnographic approach could be extended to other emerging technologies, such as biotechnology or autonomous systems, where similar neutrality myths shape education and practice.","The paper's emphasis on global perspectives suggests that AI ethics curricula designed in the Global North may need local co-creation with engineers and communities in the Global South to avoid replicating the very exclusions the authors identify."],"forward_implications":["AI engineering curricula would integrate ethics, social responsibility, and interdisciplinary collaboration as core components rather than optional add-ons.","Engineering programs would teach transparency as an ethical design principle, not just code documentation.","Graduates would be expected to question technological neutrality and technosaviourism, recognizing AI systems as sociotechnical artifacts.","Curricula would need to incorporate global and culturally responsive perspectives, bridging the gap between academia and industry.","AI education would aim to cultivate ethical leaders with lifelong learning mindsets, not merely proficient coders."],"supporting_citations":[{"why":"Supplies the critique of conventional engineering education that underplays emotions and calls for integrating emotional intelligence, empathy, and social responsibility.","marker":"[2]"},{"why":"Provides the whole-person education model centered on moral motivation and the powers of mind, heart, and will that the paper extends to AI education.","marker":"[5]"},{"why":"Introduces value-sensitive design as a framework for embedding human values into AI system development, a key recommended curricular element.","marker":"[8]"},{"why":"Offers a concrete case study of racial bias in facial recognition that illustrates AI's non-neutrality and the societal stakes of engineering decisions.","marker":"[17]"},{"why":"Deconstructs AI's myth of neutrality by tracing its ecological and labor costs, grounding the paper's challenge to technological neutrality.","marker":"[20]"},{"why":"Provides the inclusive design framework's three dimensions that the paper argues should be embedded in engineering education.","marker":"[29]"},{"why":"Articulates sustainable development as a meta-context for engineering curriculum renewal, supporting the call for a holistic, system-wide lens.","marker":"[42]"},{"why":"Defines collaborative autoethnography as the method used to collect and interpret participant reflections in this study.","marker":"[45]"},{"why":"Supplies the six-phase thematic analysis procedure used to code and generate themes from the autoethnographic narratives.","marker":"[49]"}],"fun_headline_variants":["AI engineers need ethics, not just code","Study: AI education must include ethics and global views","Whole-person approach needed for AI engineering education","Interdisciplinary AI education: a call for whole-person skills"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The findings rest on the assumption that the autoethnographic reflections of twenty self-selected advocates for whole-person education are representative of what AI engineering education broadly needs; if those reflections are not generalizable, the prescriptive force of the argument weakens.","fun_headline_variants_meta":{"raw":{"variants":["AI engineers need ethics, not just code","Study: AI education must include ethics and global views","Whole-person approach needed for AI engineering education","Interdisciplinary AI education: a call for whole-person skills"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000336,"raw_usage":{"total_tokens":1802,"prompt_tokens":826,"completion_tokens":976,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":442,"completion_tokens_details":{"reasoning_tokens":916}},"tokens_in":442,"tokens_out":976,"duration_ms":8143,"temperature":1.0,"reasoning_tokens":916,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T04:54:27.932431+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A comparative study tracking AI engineering graduates from whole-person-integrated programs versus traditional technical programs could test the claim: if both groups show equivalent ethical reasoning, interdisciplinary collaboration, and societal impact in their work, the paper's central premise would not hold. Alternatively, a large-scale survey of AI engineering educators and practitioners asking whether they see the same gaps in current curricula would provide evidence for or against the universality of the reported motivations.","supporting_citations":[{"cited_title":"Awakening engineering education,","cited_arxiv_id":null,"evidence_quote":"Supplies the critique of conventional engineering education that underplays emotions and calls for integrating emotional intelligence, empathy, and social responsibility."},{"cited_title":"A whole-person approach to educating for sustainability: Developing a values-based and systemic approach to transformative learning,","cited_arxiv_id":null,"evidence_quote":"Provides the whole-person education model centered on moral motivation and the powers of mind, heart, and will that the paper extends to AI education."},{"cited_title":"Mapping value sensitive design onto ai for social good principles,","cited_arxiv_id":null,"evidence_quote":"Introduces value-sensitive design as a framework for embedding human values into AI system development, a key recommended curricular element."},{"cited_title":"Gender shades: Intersectional accuracy disparities in commercial gender classification,","cited_arxiv_id":null,"evidence_quote":"Offers a concrete case study of racial bias in facial recognition that illustrates AI's non-neutrality and the societal stakes of engineering decisions."},{"cited_title":"Anatomy of an ai system: The amazon echo as an anatomical map of human labor, data and planetary resources,","cited_arxiv_id":null,"evidence_quote":"Deconstructs AI's myth of neutrality by tracing its ecological and labor costs, grounding the paper's challenge to technological neutrality."},{"cited_title":"The value of being different,","cited_arxiv_id":null,"evidence_quote":"Provides the inclusive design framework's three dimensions that the paper argues should be embedded in engineering education."},{"cited_title":"Sustainable development as a meta-context for engineering education,","cited_arxiv_id":null,"evidence_quote":"Articulates sustainable development as a meta-context for engineering curriculum renewal, supporting the call for a holistic, system-wide lens."},{"cited_title":"Collaborative autoethnography,","cited_arxiv_id":null,"evidence_quote":"Defines collaborative autoethnography as the method used to collect and interpret participant reflections in this study."}],"review_version":1}