athlete with smartphone AI coach

New Science – people prefer coaching from a trained AI vs written by human expert coach

Svexa’s Co-Founder, Chair of the Board and Chair of Stanford Medicine Prof. Euan Ashley, M.B., Ch.B., D.Phil. and svexa Co-Founder, CEO Mikael Mattsson PhD were part of the team who contributed to an interesting new scientific article that shows athletes actually prefer the coaching messages from a trained AI vs a human coach.

The full article is title “Fine-tuning Large Language Models in Behavioral Psychology for Scalable Physical Activity Coaching”, and was published in medRxiv on 21st February.

Personalized, smartphone-based coaching improves physical activity but relies on static, human-crafted messages. The research introduced My Heart Counts (MHC)-Coach, a large language model fine-tuned on the Transtheoretical Model of Change. MHC-Coach generates messages tailored to an individual’s psychology (their “stage of change”), providing personalized support to foster long-term physical activity behavior change. To evaluate MHC-Coach’s efficacy, 632 participants compared human-expert and MHC-Coach text-based interventions encouraging physical activity. Among messages matched to an individual’s stage of change, 68% preferred MHC-Coach-generated messages. Blinded behavioral science experts rated MHC-Coach messages higher than human-expert messages for perceived effectiveness at 4.4 vs. 2.8.

MHC vs Human Coach

This work demonstrates how language models can operationalize behavioral science frameworks for personalized health coaching, promoting long-term physical activity and potentially reducing cardiovascular disease risk at scale. It is somewhat unexpected for the AI coach to perform better than human experts. This provides a key learning, that AI can provide scale and still personalize to everyone without dropping in engagement or becoming boring. The optimal solution may well be to combine the ‘always available and never tired AI’ with the human connection, and have the AI as a co-pilot.

This research also validates one of svexa’s central algorithm tenets, that an AI coach built using decades of real-world experience can indeed deliver personalized, adaptive and therefore compelling messages to a large volume of customers.

Read the full scientific article

Euan Ashley

Euan Ashley is Chair of the Stanford School of Medicine, the Roger And Joelle Burnell Professor of Genomics and Precision Health, and a professor of medicine, of genetics, and of biomedical data science. Ashley’s research group is focused on the science of precision medicine. In 2010, he led the team that carried out the first clinical interpretation of a human genome. In 2023, he published the book “The Genome Odyssey: Medical Mysteries and the Incredible Quest to Solve Them.”

newsletter signup

Once in a while we send our subscribers a newsletter with corporate news, useful industry information or just a sneak peak of what we’re doing at svexa.

    Please prove you are human by selecting the key.