Colleen Caleshu, MS, CGC, on the Use of Artificial Intelligence in Genetic Counseling

“...Clinicians, including counselors, really need to get AI literate to be able to assess the appropriateness of these tools, the quality, what biases they have, whether or not they're equitable—so sort of striking a good balance of having an open mind and being curious about how you can improve your practice using them, while also being thoughtful and thinking critically about them.”

In recent years, artificial intelligence (AI) has begun to make its way into the medical field, and this has included the field of genetic counseling. At the 44th National Society of Genetic Counselors (NSGC) Annual Conference, held November 6 to 10 in Seattle, Washington, a session was held on the growing use of AI by genetic counselors entitled “Healthcare Administrative AI Tools in Practice: Implementation Insights from Early Adopters.” At the conference, CGTLive® sat down with Colleen Caleshu, MS, CGC, the senior director of research and real world data at Genome Medical, to learn more about what was discussed by the speakers at the session and the key takeaways.

Caleshu noted that in the session early adopters of AI in their clinical practice shared which AI tools they use, how they implemented them, and what worked or didn’t work. She noted that adoption has surged: many health systems now offer tools—most commonly ambient recording and transcription to autogenerate clinical notes—but many solutions aren’t tailored to genetic counseling specifically.

Caleshu highlighted clear benefits of AI in clinical practice: increased efficiency, reduced documentation burden, and potential to ease clinician burnout. But she emphasized caution: clinicians must become AI-literate to evaluate tool quality, biases, and equity implications. Validation is a major challenge—implementers should ask what validation was done, on whom, and with what results. Furthermore, most deployments remain “clinician-in-the-loop” because generative outputs can be inaccurate or inappropriate; speakers shared examples of AI producing off-base notes or patient messages. Caleshu concluded that clinician oversight remains critical at this point in time.