The chatbot will see you now

It’s right there in your pocket. The AI assistant on your phone won’t ask you to make an appointment. It won’t lose patience, and it’ll eagerly give you the attention you deserve. But would you risk consulting ChatGPT, Gemini, or any of the multitude of chatbots out there for health advice?

I did. And it was worth it.

Not long ago, I noticed a pain on the inner side of my left arm that had become dramatically worse. Being a catastrophiser, I instantly thought: heart attack. Luckily, I didn’t go so far as to call an ambulance. Instead, I consulted Grok Doc—what xAI calls one of Grok’s voice modes.

I described the pain in detail and asked if I could be having a heart attack. Grok quizzed me: How long had the pain been there? Was there bruising or swelling? Had I strained that arm in particular? A dozen questions followed. I’d had the pain for over three weeks, so Grok confidently ruled out a heart attack.

Read this | Mint Primer: Dr AI is here, but will the human touch go away?

Because Grok had asked, I looked closely and noticed there was definitely swelling. And then, in a flash, I realised what had happened—an over-enthusiastic lab technician had made my arm hurt during a recent blood test. I told Grok, and it suggested using an ice pack thrice a day and, if really needed, taking an anti-inflammatory—within dosage limits. It told me to rest the arm and come back in two or three days, for all the world as if it were really a doctor.

I’m afraid I was too lazy for the ice packs and didn’t bother with the anti-inflammatory, but I was happy to rest the arm—as well as the rest of me. The pain gradually eased.

Despite disclaimers about not replacing doctors, AI assistants will give health-related advice if you ask. That’s not what Gemini, ChatGPT, Grok and their ilk are meant for, but since they can scan reputable sources in seconds, their advice can often be helpful—if the user is careful in framing questions and responsible in following through. These tools are still prone to hallucinations, and they’re only as good as their training data, so they can get things wrong.

Read this | Mint Primer | Who is liable if a friendly chatbot ‘abets’ suicide?

Luckily, there are ways to manage that risk. You can cross-check with another AI chatbot. Ask the AI to verify anything that seems off. Even better, run it by your doctor to see how on-point the suggestions usually are.

It’s wise to be cautious, but there’s no need to throw the baby out with the bathwater. I fed ChatGPT my test reports and had it explain each parameter in detail. Take uric acid, for instance—it was within range and not flagged by the lab, but on the higher side of normal. Rather than wait for it to spike, I asked ChatGPT how I could bring it down a bit. The AI gave me dietary tips and even a workout schedule, complete with specific exercises.

Frankly, AI’s advice beat anything I’ve received from the dietician’s assistant at the doctor’s clinic, who usually hands over a generic, one-size-fits-all restrictive regime. Chat assistants, on the other hand, can offer deeply personalised advice. Share your height, weight, vulnerabilities and preferences, and you’ll get practical, tailored guidance.

Depending on the chatbot, you can even ask for a printable version of the recommendations—handy if you want to take it along to a doctor’s appointment. You can request a list of questions to ask during the visit, too.

One underrated feature of some chat assistants is the ability to go live with the camera. You can turn it on, show Gemini or ChatGPT your meal, and get information like calories, glycemic index, and more. It’s surprisingly useful—most users don’t even know this exists. You could also show the AI an injury and get suggestions, though that edges into riskier territory.

Also read | The brain behind Generative AI has his sights set on India

Needless to say, anything remotely serious should be taken to a medical professional. But for supportive advice, don’t rule out the chatbot.

Meanwhile, I know I’ll be forever wary of lab technicians.

Mala Bhargava is most often described as a ‘veteran’ writer who has contributed to several publications in India since 1995. Her domain is personal tech and she writes to simplify and demystify technology for a non-techie audience.