Why we do it
Reasons
Lack of effective governance increases the risk and hinders the adoption of Responsible AI solutions towards better health outcomes.
This lack of national governance mechanisms contributes to the slow adoption of AI solutions within health systems. Governments are hesitant to approve technologies without more substantial evidence of technology’s safety and efficacy; technology developers do not have a clear path for certification and approval from regulatory agencies; and private sector companies are left to develop ethical frameworks without the broad, inclusive network or experience needed, resulting in frameworks that may be too narrow, incomplete, or misaligned with the public good.
Strengthening the governance of AI in health is necessary to safeguard the future of health. To realize the potential of these technologies and keep citizens safe, we must ensure these technologies contribute effectively to global progress on health and well-being.
Better Well-being
Improved health and well-being outcomes for all
We have only just begun to see the implications of AI-powered healthcare. Natural language processing, learning models, and many other AI tools will continue to forge new pathways of learning, connecting, and creating that will result in new medicines, diagnostic tools, service types, and even new understandings of “health and well-being.” Paired with the vast amounts of data available in health systems today, AI can and will define the next phase of health.
AI is already contributing to drug development, radiology and imaging, outbreak monitoring, and health information dissemination. Researchers and technologists worldwide are actively working on new tools and platforms to tackle some of the most complex challenges facing health systems today.
By proactively addressing the digital divide, we can ensure the benefits of AI-powered health are shared equitably across all countries and communities, leaving no one behind. By constructing strong, responsive regulatory systems, we can preemptively address the risks and harms that AI can cause.
What is ‘Responsible AI’?
The term ‘Responsible AI’ refers to artificial intelligence technologies that align with requirements set by normative agencies and other sector leaders, with a specific focus on ethical, human-centric attributes. HealthAI generally defines ‘Responsible AI’ as:
AI solutions that are ethical, inclusive, rights-respecting, and sustainable.
Attributes of Responsible AI include:
Protection of and respect for human autonomy, agency, and oversight
Promotion of human well-being and safety
Commitment to “do no harm”
Technical robustness and safety
Adherence to laws and ethics
Transparency, Explainability, and Intelligibility
Responsibility and accountability
Inclusivity and equity
Sustainability
Societal and environmental well-being.
These attributes are applied across all aspects of AI technologies, from the technical development of the solution to the use and management of data, the implementation and stewardship of the technology, uses of the technology, and the ultimate result of its application.
This definition is derived from the WHO publication titled Ethics and Governance of Artificial Intelligence for Health, the International Development Research Center’s AI for Global Health Initiative, a framework developed by the European Commission’s High-Level Expert Group on AI, described in the Ethics Guidelines for Trustworthy Artificial Intelligence, and a journal publication from Information Systems Frontier.
Our Impact
Our work creates the trust, equity, and sustainability required to achieve the full potential of Responsible AI for health.
Our work creates the trust, equity, and sustainability required to achieve the full potential of Responsible AI for health. It contributes to improved health and well-being outcomes for all in alignment with the Sustainable Development Goals.
Increased Access to Safe, High-Quality, Effective, and Equitable AI Solutions.
- Ensure AI solutions are safe for use, comply with quality standards, and effectively deliver their intended health outcome or system improvement.
- Provide information on market access authorization and reimbursement processes.
- Support an early-warning mechanism to alert countries of adverse events.
- Streamline information sharing between countries to propagate the availability of proven Responsible AI solutions.
Increased Trust, Investment, and Innovation in Responsible AI Solutions for Health.
- Protect national data sovereignty and ensure that all real-world data is collected, shared, and used under regulatory rules and through approved data centers-
- Ensure compliance with internationally defined Responsible AI standards.
- Support validation processes that account for the usage of real-world data and enable feedback from civil society;
- Foster an ecosystem that promotes investment in the research, development, and adoption of Responsible AI solutions for health.
Increased Government Revenue From Regulatory Activities.
- Generate new sources of revenue for regulatory agencies and government budgets to allow for sustained funding for the regulatory mechanisms and additional investment capacity.
- Accelerate approval processes across countries, leading to cost savings and bureaucratic streamlining.
By supporting country-driven regulatory mechanisms, we promote safe, effective, and high-quality AI technologies that improve health, reduce costs, and expand the reach of health services.
Strong, responsive regulation enhances trust in AI technologies so that policymakers, health workers, and patients alike are confident in the efficacy and ethics of these tools. We help countries design sustainable regulatory systems and provide new revenue sources to support these essential activities.