AI is quickly weaving itself into the fabric of healthcare — but formal guardrails are still missing.
While Congress debates, a coalition of clinics, tech firms, and hospitals is stepping up with its own playbook.
Their goal: build trust, ensure fairness, and prevent the chaos of unregulated AI in medicine.
Inside the Movement to Regulate Health AI
The U.S. government may not have clear rules for how AI is used in healthcare, but that hasn’t slowed the tech’s rollout. From automated scribes to AI-powered agents, tools are already being tested in hospitals and clinics across the country.
A recent survey from the Deloitte Center for Health Solutions found that 75% of leading healthcare organizations are experimenting with generative AI. Even more telling: 82% are actively setting up internal oversight and governance systems.
That’s where the Coalition for Health AI (CHAI) enters the picture. Founded in 2021, CHAI is a public-private effort that now includes over 3,000 stakeholders — from the Mayo Clinic and Cleveland Clinic to smaller startups — all working together to shape responsible AI use in healthcare.
CHAI has already produced a 180-page framework packed with practical standards. Released in June 2023, it covers everything from transparent system design to fairness in training models. The guide isn’t legally binding, but it’s fast becoming a go-to resource.
At the center of CHAI’s approach is a tool called the “model card.” Think of it as a nutrition label for AI tools. It spells out key metrics, performance benchmarks, and ethical considerations. If a health system rolls out an AI scribe, for example, the model card would explain how it works, what biases may exist, and how it was tested for safety and effectiveness.
CHAI isn’t lobbying for sweeping new laws. Instead, the group wants to guide industry best practices and provide public officials with grounded insights from the frontlines of AI deployment.
That work now includes a fresh partnership. On June 11, CHAI joined forces with the Joint Commission — the nonprofit that accredits U.S. hospitals — to create new AI safety certifications.
According to the release, their joint effort will debut a new guidance framework this fall, with accreditation programs to follow.
What’s at Stake for Patients and Providers
Without national AI laws in healthcare, states are left to set their own rules — or none at all. That patchwork approach could confuse providers, slow adoption, or even harm patients.
For example, AI that performs well on urban datasets may struggle in rural communities. That’s why fairness isn’t just a nice-to-have for CHAI — it’s essential. Their standards emphasize inclusivity, urging developers to test tools across diverse populations, from tribal health centers to major urban hospitals.
The group is also prioritizing access. Many community clinics lack the budgets and bandwidth of academic medical centers. CHAI says it’s actively working on “playbooks” tailored for these smaller providers, offering step-by-step guidance on adopting AI responsibly.
The stakes are real. With AI increasingly involved in everything from diagnostics to billing, even small missteps could lead to misdiagnoses or discrimination. CHAI’s framework aims to reduce those risks before they escalate.
Insight from the Front Lines
“Transparency is critical,” said CHAI CEO Brian Anderson in an interview with Healthcare Brew.
“We want to provide the kind of guardrails and guidelines that build trust across the spectrum — from payers to providers to patients.”
Anderson added that CHAI isn’t pushing for federal policy but hopes to be a trusted educational partner for regulators.
Another member, Emmanuel Oquendo, CEO of Puerto Rico–based BrainHi, joined CHAI to help ensure AI models reflect the needs of Latino communities. “Some algorithms might work great for a Caucasian population but not others,” he noted.
Oquendo is now part of a CHAI working group developing standards for AI-powered healthcare chatbots.
Where It Could Go From Here
CHAI is quietly becoming the playbook writer in a field with no rulebook. With Congress still hesitating on national AI policy for healthcare, this coalition may be the most influential body shaping how hospitals and health startups deploy these tools.
The upcoming accreditation framework could also raise the bar across the industry — especially if insurers or major providers begin to require it.
In the absence of federal law, CHAI’s voluntary standards might just become the de facto ones.
What Do You Think?
Should healthcare AI follow voluntary standards or wait for government rules? Tell us what matters most to you.
About Author:
Eli Grid is a technology journalist covering the intersection of artificial intelligence, policy, and innovation. With a background in computational linguistics and over a decade of experience reporting on AI research and global tech strategy, Eli is known for his investigative features and clear, data-informed analysis. His reporting bridges the gap between technical breakthroughs and their real-world implications bringing readers timely, insightful stories from the front lines of the AI revolution. Eli’s work has been featured in leading tech outlets and cited by academic and policy institutions worldwide.
Pingback: AI “Therapists” Are Talking Back — But At What Cost?