Most CME/CE and accreditation teams are already using AI. It is drafting needs-assessment summaries, clustering evaluation comments, generating assessment distractors, and adapting materials for different audiences. In 2026, the real question is whether the way your team uses these tools holds up to an accreditation review.
On January 30, 2026, ACCME, together with Joint Accreditation, released Guidance on the Responsible Use of Artificial Intelligence in Accredited Continuing Education. It is not a new rule set so much as a clarification: the Standards for Integrity and Independence still apply, AI or not. What changes is where the work lives. Independence, disclosure, content validity, and learner privacy now have to be managed inside your AI workflows, not around them.
For directors, that reframes the problem. This is not a tooling decision your team makes once. It is an infrastructure question: who reviews AI-assisted content, how AI use is disclosed, where outputs are stored, and how all of that is documented well enough to show an accreditation reviewer. Below is a plain-language read of what the guidance asks for, organized the way the guidance itself is.
The guidance encourages AI, within the Standards
The first thing to know is that ACCME and Joint Accreditation are not discouraging AI use. The guidance explicitly invites providers to experiment with it for needs assessments, content drafting, assessment design, evaluation analysis, and case-building.
The encouragement comes with a condition that runs through every section: AI augments professional judgment; it does not replace it. The accredited provider remains fully responsible for the content.
What the guidance asks of accredited providers
1. Safeguard independence and mitigate bias
AI-generated or AI-assisted content must meet the same independence bar as anything a human author produces. Content stays grounded in current science and presents a balanced view of diagnostic and therapeutic options. Outputs are screened for commercial bias or promotional language, and people without relevant financial relationships make the decisions about how AI is used.
No advertiser, supporter, or ineligible company may influence an AI recommendation engine used in an accredited activity, and providers should document the steps taken to keep AI content free of commercial influence.
2. Transparently disclose AI use
When AI is used to generate, modify, or analyze educational materials, that use should be disclosed. The tool name, version, and date, the purpose it served, and confirmation that a human reviewed the output. Routine spelling and grammar tools are exempt.
Disclosure here is an extension of the same informed-engagement principle that underpins financial-relationship disclosure: learners should know what shaped the education in front of them.
3. Ensure human oversight, accuracy, and accountability
Static AI-generated materials — slides, handouts, and assessments — must be reviewed and approved by named, qualified individuals before they reach learners, checked for fabricated citations and “hallucinations,” screened for bias in clinical or demographic representation, and carried with version control that records who reviewed what and when.
Where learners interact with AI live, the system itself needs clinician oversight, and learners are reminded that final clinical responsibility always rests with the professional.
4. Protect learner identity and sensitive information
Don’t put protected health information or personally identifiable information into tools that aren’t approved for it. Get consent before any learner data or identity is shared with a third party, de-identify outcomes data beyond internal use, and keep data governance and retention aligned with institutional and legal requirements.
5. Limit prohibited or high-risk uses
Some uses call for especially rigorous oversight or should be avoided. The guidance flags generating diagnostic or treatment recommendations without clinical validation, storing sensitive content in public tools, auto-producing assessment answers visible to learners, and automating summaries that bypass bias, independence, or accuracy checks.
6. Establish internal governance and continuous improvement
For any provider using AI at scale, the guidance points to governance: a named owner for AI use, an approved-tools list, role-specific policies for staff, faculty, and learners, piloting before full rollout, and usage logging with periodic review. This is the part that turns scattered, individual AI use into something an accreditation reviewer can actually see and trust.
Structuring all of this from scratch is the hard part. Our new AI in Accredited Education: A Governance Checklist lays each of these seven areas out as a working checklist. Get the Checklist →
7. Secure databases and AI systems
Confidential, proprietary, or unpublished material goes only into private or closed systems, with faculty permission where their content is involved. Store AI-assisted content and learner data in secure, access-controlled environments aligned to HIPAA, FERPA, and GDPR, and audit routinely to confirm nothing AI touched has drifted out of compliance.
The through-line: this is an infrastructure problem
Read together, the seven areas describe a system, not a checklist of one-off tasks. Disclosure language has to live in your content templates. Review and approval need named owners and version history. An approved-tools list and a governance policy have to exist before scale, not after. Storage has to be secure and documented. None of this is hard in principle, but it is steady, cross-functional work that lands on teams already running a full slate of activities.
That is the gap most internal CME/CE teams feel right now. The expertise to evaluate AI outputs is clinical, the workflow to govern them is operational, and the documentation to prove it is administrative. Keeping all three coherent across every activity is exactly the kind of load that doesn’t scale by adding it to existing roles.
Where CineMed fits
CineMed CE operates as your external CME/CE office: the education infrastructure behind your accredited activities. That means audit-ready documentation, disclosure and review workflows, and joint accreditation expertise built into how activities are planned and run, so adopting AI doesn’t mean improvising new compliance processes under deadline.
The objective isn’t to slow your team down. It’s to let you use AI confidently, with the governance and recordkeeping already handled.
If you want a practical starting point, our new AI in Accredited Education: A Governance Checklist walks through each of these seven areas as a working checklist your team can apply this quarter.
Get the Checklist
AI in Accredited Education: A Governance Checklist — a working checklist that walks through all seven areas ACCME’s guidance covers, so your team can pressure-test its AI workflows this quarter.
See What Audit-Ready Looks Like
Our CME Audit-Readiness Toolkit: a full audit prep checklist, the compliance gaps teams miss most, and a ready-to-use CME Activity Tracker.
See What Audit-Ready Looks Like →
Talk to a CE Specialist
If AI is stretching your existing compliance processes thin, let’s talk through what running it on audit-ready infrastructure would look like.
In support of improving patient care, CineMed is jointly accredited by the Accreditation Council for Continuing Medical Education (ACCME), the Accreditation Council for Pharmacy Education (ACPE), and the American Nurses Credentialing Center (ANCC), to provide continuing education for the healthcare team.
