Objective: Develop and promote responsible AI standards in healthcare to ensure high-quality, trustworthy, and equitable AI applications.

This includes:

• Building trust: Ensuring that AI in healthcare is reliable and widely accepted and understood by patients, clinicians, and organizations.
• Broad representation: Engaging diverse stakeholders, including underserved communities, to ensure AI tools are developed and tested across various environments and data sets.
• Public/private collaboration: Working with both the public and private sectors, leveraging the strengths of each to develop effective AI standards and tools.
• Transparency: Promoting transparency in AI development and implementation, addressing concerns about the “black box” nature of AI decision-making.
• Supporting innovation: Balancing the need for regulation with the need to foster innovation by developing consensus-driven standards to ultimately increase adoption of safe and effective AI.

CHAI DELIVERABLES

CHAI aims to create a federated network of labs and a robust framework that can be used by all healthcare providers, including those in under-resourced settings, to validate AI models and ensure they benefit all patient populations.

We will do this by:

• Certification of Assurance Labs: Supporting the development and certification of a nationwide network of quality assurance labs to evaluate and validate AI models using community best practices and standards.
• Continuous evaluation: Providing tools and frameworks for ongoing testing and monitoring of AI models to ensure they remain safe, effective, and unbiased over time.
• Best Practice Guidance: With the input of experts that have real-world experience, developing best practice guidance for developing and deploying health AI technologies demonstrated use case by use case.
• National Registry: To support greater transparency, CHAI will publish model report cards on a publicly-available registry.