Physical Therapists

Physical therapy is a very hands-on field, so most core tasks still need a human. For example, therapists evaluate patients and design custom exercise programs, then teach those exercises to patients and families ^{[1] [1]}. They also measure strength or balance and adjust treatments by feel, and they explain the risks and benefits of therapy in person ^{[1] [1]}.

Some routine parts can use AI or software help: for instance, speech-to-text programs can draft progress notes, and apps or sensors can count reps or track simple exercises. But even those are just tools for the therapist. In fact, official data rate the job as only about 14% automated today ^[1], meaning most of the work is done by people.

In short, AI can help document data or send reminders, but the actual exam, hands-on exercises, and personal coaching are done by therapists.

New AI tools for PT are still rare, so adoption will likely be gradual. One reason is cost: clinics pay therapists about $97,960 a year on average ^[2], so any new technology must clearly save money or improve care to be worth buying. Also, physical therapy is regulated healthcare.

Tasks like getting a patient’s informed consent or developing a rehab plan must be done personally by a licensed therapist ^[1]. Patients and doctors often prefer a real person guiding them through exercises and safety rules. On the other hand, there are advantages: for example, AI-driven scheduling or smart exercise reminders could free up therapist time, and tele-rehabilitation tools (using video or apps) are growing.

Overall, AI is more likely to augment PT work (helping with charts or follow-up) than replace it. Therapists’ judgment, empathy, and physical skills remain central, so technology will probably be used as a support — giving therapists more time for the human parts of care ^{[1] [2]}.