{"resourceId":"hando-msse-ai-not-ready-sports-medicine-2026","versions":[{"version":"external-4420b4e91f2ce7547b8d2794ca4f5cc7cdc78eb343b812a142bf19971bf1fa5c","resource":{"id":"hando-msse-ai-not-ready-sports-medicine-2026","title":"Hando et al. MSSE response: current sports-medicine AI/ML lacks real-world validation for clinical use","organization":"Medicine & Science in Sports & Exercise (ACSM)","sector":"Sports medicine / exercise science (athlete and military populations)","geography":"United States (UVA / Baylor–Army authors; military and athletic deployment contexts)","publishedAt":"Publish Ahead of Print (MSSE 2026); discussed in Newswise 2026-06-23","publicationDate":null,"eventDate":null,"sourceName":"Medicine & Science in Sports & Exercise","sourceLabel":"Academic response article — AI readiness in sports medicine","sourceUrl":"https://www.newswise.com/pdf_docs/17821565104422_Response_artificial_intelligence_is_not_a_useful_tool_in.1089.pdf","evidenceClass":"academic-research","outcomeClass":"cautionary","topics":["governance-procurement","data-security","developers-agents","operating-model"],"finding":"In an MSSE ahead-of-print response (DOI 10.1249/MSS.0000000000004046), Hando, Kupperman, Brown, and Angadi argue current AI/ML systems used by athletes, military units, and health organizations have not undergone requisite real-world testing; where externally validated, outputs were inaccurate or misleading. They stress transparency/explainability/accountability, criticize black-box wearables and circular validation, note high LLM hallucination rates, and conclude evidence does not support clinical AI/ML use in sports medicine pending rigorous research. Table contrasts optimistic claims with limited study utility (e.g., injury model true-positive rate 53.7% without prospective validation).","sledRelevance":"Why now: Directly counters the deployment narrative around collegiate injury/readiness ML highlighted by the 2026-09-28 BMC high-accuracy Kaggle study and ongoing vendor injury-AI marketing. Provides peer-reviewed sports-medicine scrutiny athletics clinicians can cite in procurement gates.","evidence":"Full PDF inspected: authors state disagreement that AI has arrived for clinical use; cite their prior military cohort validations showing poor utility; discuss black-box circular validation; GPT-5 hallucination range 26–47% cited; Table 1 critiques imaging/injury/digital-twin claims; recommend external validation, adversarial testing, monitoring. Military-focused empirical base; collegiate use referenced as deployment context in related coverage.","architectureImplications":"Do not wire unvalidated injury-risk or readiness models into return-to-play decision systems; prefer transparent models with prospective monitoring and human clinical authority.","governanceImplications":"Apply SaMD-like expectations (transparency, real-world performance monitoring, pre-use validation) to sports-science software affecting athlete health decisions.","securityPrivacyImplications":"Black-box vendor models processing wearable/biomechanical data impede auditability; contracts should require explainability and independent evaluation rights.","caveats":"Response essay, not a new multi-school RCT; strongest empirical citations are military cohorts; exact print date not on PDF (ahead of print); collegiate-specific metrics limited.","streamIds":["college-athletics"],"roles":{"sales":"Sales framing: Peer-reviewed sports-medicine authors say unvalidated AI in athlete/military settings is a liability, not an edge—especially injury prediction sold without external tests. Open clinical/performance deals with validation evidence requirements and shadow-mode pilots. Do not lead with AI-has-arrived messaging. Unsupported: that all wearables are useless, or that future validated AI cannot help; this source rejects premature clinical deployment only.","engineering":"Engineering framing: Fit for athletics is decision-support only after external validation and ongoing monitoring; architecture must expose features, decision-time data, and override. Prerequisites: prospective label collection and independent test cohorts. Constraints: noisy wearables, circular device-as-ground-truth, LLM hallucination if used for literature tools. Security: auditable model versions. Proof of value: prospective sensitivity/specificity and calibration—not vendor AUC from internal fits.","delivery":"Delivery framing: Inventory any injury/readiness AI in athletics care, quarantine unvalidated scores from RTP decisions, and commission external validation or discontinue. Dependencies: sports medicine, ATC, counsel. Ownership: head team physician / sports medicine lead. Skills: critical appraisal of ML claims. Governance: SaMD-inspired checklist before purchase. Acceptance: written ban on unverified scores affecting availability. Risks: false security and missed prevention from acting on noise."},"roleDetails":{"sales":{"customerProblem":"Vendors sell injury-risk and readiness AI to colleges without the external validation sports-medicine reviewers say is required for clinical use.","stakeholders":["Team physicians","ATCs","Sports performance","AD / risk management"],"discoveryQuestions":["Has the tool been prospectively validated in a similar population?","What happens when the score disagrees with the clinician?","Can we independently audit the model?"],"valueHypothesis":"Validation-first sports-medicine AI advisory reduces clinical and reputational risk versus buying opaque injury predictors on marketing claims.","potentialEngagement":"Sports-medicine AI inventory and evidence review against SaMD-like criteria for one high-risk tool.","unsupportedClaims":["Guaranteed injury prevention","That military validation results transfer unchanged to every NCAA sport","That LLMs safely generate physiological hypotheses"]},"engineering":{"fit":"Clinical decision-support governance for athlete health AI; reject black-box injury oracles without prospective proof.","architectureIntegration":"Shadow scoring with clinician override; model cards; monitoring for drift; no closed-loop RTP automation.","prerequisites":["Prospective outcomes pipeline","Independent validation cohort plan","Explainability requirements in vendor RFP"],"constraints":["Noisy wearable ground truth","Rare injuries","Proprietary vendor refusal to open models"],"security":"Audit logs for score-influenced decisions; prohibit undisclosed secondary model training on athlete data.","proofOfValue":["Prospective sensitivity/specificity","Calibration plots","Adversarial/external test results"]},"delivery":{"work":"Pause clinical reliance on unvalidated tools; design prospective evaluation or decommission; train clinicians on failure modes.","dependencies":["Sports medicine authority","Vendor data access for audit","IRB/privacy if researchifying evaluation"],"ownership":"Head team physician / sports medicine medical director.","skillsAdoption":"Clinician literacy on external validation vs marketing AUC.","governanceCheckpoints":["Purchase gate with evidence dossier","Quarterly performance monitoring","Incident review when scores mislead"],"acceptanceCriteria":["No unvalidated score as sole RTP determinant","Documented external validation or tool removal"],"risks":["False reassurance","Unnecessary restriction of athletes","Liability from opaque recommendations"]}},"retrievedAt":"2026-10-01T03:19:55Z","enrichedAt":"2026-10-01T03:27:55Z","enrichmentBasis":"retrieved source","accessibilityWorkforceImplications":"Keeps clinicians and experienced practitioners central; warns humans-with-unvalidated-AI can be worse than trained humans without it.","procurementImplications":"RFPs for injury/readiness software should demand peer-reviewed external validation, adversarial testing, and real-world monitoring—not utilization rates.","operatingModelImplications":"Sports medicine should treat most commercial AI as experimental until prospectively proven, with clear clinical ownership.","sourceVerification":{"responseId":"curl-msse-pdf-20261001T031955","openedUrl":"https://www.newswise.com/pdf_docs/17821565104422_Response_artificial_intelligence_is_not_a_useful_tool_in.1089.pdf","referenceExcerpt":"the current AI/ML systems in use by athletes, military units, and health organizations have not undergone the requisite testing","promptVersion":"sled-research-v4.0","model":null,"basis":"agent-reported inspection"}}}]}