AI Inquiry Atlas

An unofficial, independent guide to the public records of the Joint Select Committee on Artificial Intelligence. Not connected with the Parliament of Australia or the committee. Its makers also made a submission (no. 481; see About). Summaries and scores are written by AI and can be wrong: check the original before relying on them.

Submission 118

Royal Australasian College of Surgeons

Submitter group: Health.

An AI model identified the submitter group and wrote the summary, key points and recommendation summaries. Counts do not measure importance or merit.

Open in the interactive guide

Summary

Read the original submission for the full text.

Argues that AI offers significant opportunities for surgical care and the health system, with nationally consistent risk-based governance, sovereign capability, education, transparency, accountability and equity safeguards. Says digital innovation must complement regional healthcare investment and preserve human judgement and professional responsibility.

Key points

Quotes are the submitter’s own words. Text taken from documents, including scanned pages read by text-recognition software, can differ from the PDF. Check the original. Listed in the same order as on the submission’s page in the interactive guide.

  1. Says it supports risk-based regulation that enables innovation with safeguards proportionate to potential harms.

    RACS supports a risk-based regulatory framework that enables innovation while maintaining safeguards proportionate to potential harms.

  2. Says Australia could build sovereign healthcare AI capability through domain-specific models informed by high-quality Australian clinical data and expertise.

    Australia has an opportunity to build sovereign capability in healthcare AI, particularly through the development of domain-specific models informed by high-quality Australian clinical data and healthcare expertise.

  3. Says clinicians will need competencies to understand AI systems, evaluate outputs, recognise limitations and exercise oversight.

    Surgeons and other health professionals will require new competencies to understand AI systems, evaluate their outputs, recognise limitations and exercise appropriate oversight.

  4. States that AI tools could help address geographic disparities in health outcomes, but that digital innovation must complement rather than replace regional investment.

    The ability to support clinical decision-making and facilitate more efficient pathways to care could help address longstanding geographic disparities in health outcomes. However, RACS emphasises that digital innovation must complement, rather than replace, investment in healthcare infrastructure, workforce capability and patient-centred services in regional Australia.

  5. Says accountability for AI use should be explicit and traceable, with surgeons accountable for their clinical judgement and decisions.

    Accountability for AI use must remain explicit and traceable. Surgeons remain accountable for their clinical judgement and decisions.

  6. Recommends informing patients when AI materially influences care, with disclosure, explanation and consent proportionate to its significance and risk.

    Patients should be informed when AI materially influences their care. Disclosure, explanation and consent processes should be proportionate to the significance and risk of AI involvement.

Recommendations

Recommendations are shown beside the submission’s own words. Listed in the same order as on its page in the interactive guide.

  1. Support a nationally consistent, risk-based AI governance framework that promotes innovation while protecting patients and the public.

    Support a nationally consistent, risk-based framework for AI governance that promotes innovation while protecting patients and the public.

  2. Invest in sovereign Australian health and medical AI capability, including domain-specific models informed by Australian data and expertise.

    Invest in sovereign Australian health and medical AI capability, including domain-specific AI models informed by Australian data and expertise.

  3. Strengthen requirements for evaluation, monitoring and post-deployment surveillance of AI systems used in healthcare.

    Strengthen requirements for evaluation, monitoring and post-deployment surveillance of AI systems used in healthcare.

  4. Promote transparency and accountability standards that preserve patient trust and professional responsibility.

    Promote transparency and accountability standards that preserve patient trust and professional responsibility.

  5. Embed privacy, security and Indigenous data sovereignty considerations in all AI-related policy development.

    Embed privacy, security and Indigenous data sovereignty considerations in all AI-related policy development.

  6. Invest in AI literacy, education and workforce capability across the health sector.

    Invest in AI literacy, education and workforce capability across the health sector.

  7. Ensure equity, cultural safety and inclusion are core considerations in AI design, implementation and governance.

    Ensure that equity, cultural safety and inclusion are core considerations in AI design, implementation and governance.

  8. Support collaborative partnerships between clinicians, researchers, consumers, industry and government to guide responsible AI adoption.

    Support collaborative partnerships between clinicians, researchers, consumers, industry and government to guide responsible AI adoption.

  9. Ensure digital innovation complements rather than replaces regional healthcare infrastructure, workforce capability and patient-centred services.

    However, RACS emphasises that digital innovation must complement, rather than replace, investment in healthcare infrastructure, workforce capability and patient-centred services in regional Australia.

  10. Invest in health-focused AI research, evaluation and implementation science to support clinically effective, safe innovations applicable to Australian healthcare.

    RACS supports investment in health-focused AI research, evaluation and implementation science to ensure that innovations are clinically effective, safe and applicable to Australian healthcare settings.

  11. Ensure health-data governance provides transparent provenance, appropriate consent, privacy protections and clear accountability for secondary and commercial uses.

    Governance frameworks should ensure transparent data provenance, appropriate consent arrangements, privacy protections and clear accountability for secondary and commercial uses of health data.

  12. Have the Committee consider health-profession AI literacy, curricular integration, ongoing development in AI governance, ethics and risk management, and investment in multidisciplinary clinician, researcher, engineer and data scientist collaboration.

    The Committee should consider the need for: • AI literacy across healthcare professions. • Integration of AI capability into health professional education and training. • Ongoing professional development in AI governance, ethics and risk management. • Investment in multidisciplinary collaboration between clinicians, researchers, engineers and data scientists.

  13. Evaluate healthcare AI for its intended patient population, clinical setting and purpose; regulatory approval alone should not establish suitability for local implementation.

    Regulatory approval should not by itself be considered sufficient evidence of suitability for local implementation. AI systems should be evaluated for the patient population, clinical setting and intended purpose in which they will be used.

  14. Monitor performance degradation, bias, error and unintended consequences, with clear pathways to report, investigate and learn from failures.

    AI systems should be subject to ongoing monitoring for performance degradation, bias, error and unintended consequences. Clear pathways should exist for reporting, investigating and learning from failures.

  15. Inform patients when AI materially influences care; make disclosure, explanation and consent proportionate to significance and risk, with human review where appropriate.

    Patients should be informed when AI materially influences their care. Disclosure, explanation and consent processes should be proportionate to the significance and risk of AI involvement. Patients should have access to human review where appropriate.

  16. Keep accountability explicit and traceable, retaining surgeons’ clinical responsibility while distributing responsibility appropriately across developers, vendors, procurers and healthcare organisations.

    Accountability for AI use must remain explicit and traceable. Surgeons remain accountable for their clinical judgement and decisions. However, accountability should also be distributed appropriately across developers, vendors, procurers and healthcare organisations. AI must not create uncertainty regarding responsibility for outcomes.

  17. Consider language, disability, geography, cultural context and socioeconomic factors in AI development and implementation; meaningfully engage Aboriginal and Torres Strait Islander peoples and Māori where their health outcomes or data are affected.

    AI systems should reduce rather than reinforce existing inequities. Their development and implementation should consider language, disability, geography, cultural context and socioeconomic factors. Meaningful engagement with Aboriginal and Torres Strait Islander peoples and Māori communities is essential where AI affects health outcomes or data relating to those communities.

  18. Focus governance on identifying, mitigating and continually monitoring healthcare AI risks rather than regulating solely by technical architecture.

    RACS considers that effective governance should focus on identifying, mitigating and continuously monitoring these risks rather than attempting to regulate technologies solely based on their technical architecture.

Read the original

Open in the interactive guide