Behavioral Health Organizations Are Adopting AI Faster Than They Train Clinicians to Use It
Behavioral health organizations are deploying generative AI and large language models into clinical workflows without matching investments in staff training, creating a widening skills gap that could affect care quality and retention.
The American Psychological Association found that 56% of behavioral health practitioners have used AI in the past year-a 29% increase from the previous year. One in three use it monthly. Meanwhile, AI for healthcare continues to advance faster than organizations can train their workforces.
The computational efficiency of large language models has doubled every five months since 2020. This pace of improvement outstrips most training programs. Organizations adopting these tools now may face bottlenecks, judgment errors, and staff retention problems within three to five years when the gap between experienced and newer clinicians becomes pronounced.
The Experience Problem
Experienced clinicians can evaluate AI-generated treatment plans and documentation against their own clinical judgment. They notice gaps. A newly hired therapist without years of practice won't catch those same errors.
"I think the biggest gap is going to come in three to five years from now when a lot of experienced providers in this space are going to be able to look at output from AI to then factor in their experience from their lens to say, 'This is good enough' or 'Actually, there are all these things that I noticed,'" said Isamu Pant, founder of The Behavioral Fix, an operational consulting practice for behavioral health.
Training itself isn't enough. Real clinical work is messier and more complicated than simulations.
Organizations Taking Action
Centerstone, one of the largest nonprofit behavioral health providers in the U.S., has partnered with Lyssn, an AI platform that trains clinicians through simulated client sessions. Therapists earn continuing education credits while practicing different therapeutic approaches and receiving immediate feedback.
Bryan Baucom, Centerstone's chief information officer, said organizations that use AI to reduce administrative burden-not add it-will win on recruitment and retention. Clinicians choose jobs where they spend more time with patients, not more time on paperwork.
"If a company isn't trying to get their clinicians at a spot where their life can be measurably easier, it's going to be very simple," Baucom said. "If I've got two job offers and this company doesn't do anything with AI and this other company does and I can spend more time with my patients, which is why I'm in this field in the first place, they'll choose that."
Smaller Organizations Can Move Faster
Smaller behavioral health providers may have an advantage. They can adopt AI tools and training without the operational friction that comes with rolling out new systems across large organizations.
Pant said smaller organizations should still begin thinking about AI training now. "I actually think the smaller orgs might be better suited if they are small enough to be nimble, because adding AI into part of a larger organization is just kind of spreading a new way of operating."
Vendors often accommodate smaller groups on pricing. The time savings from AI tools can offset costs quickly.
Clinician Concerns Persist
Some clinicians worry AI will either replace them or create more work. In March, more than 2,400 therapists at Kaiser Permanente in Northern California held a one-day strike over concerns about the company's use of AI and its effect on care quality.
Paul Kim, founder and CEO of Sensible Care, a virtual mental health service, said clinician replacement isn't realistic soon. "When it comes to psychiatry and behavioral health, a 99% solution is not going to cut it."
Regulatory requirements and the complexity of clinical judgment mean licensed providers remain essential. The real risk is different: organizations that don't invest in training now may see declining care quality, compliance problems, and funding losses.
Pant said the path forward requires proactive training. "If the industry takes a proactive approach to AI training within organizations now to prepare the workforce for the future, the industry will land on its feet."
Your membership also unlocks:
