Inside: I Asked AI About My Child’s Medicine and Here Is What I Learned. Collaborative post.
It was a little after 11 at night. My youngest was hot, grumpy and wide awake, and I was standing in the kitchen holding a bottle of fever medicine I had bought months ago.
The label was tiny. The dosing chart went by weight, not age. And I genuinely could not remember whether it had been four hours or three since the last dose.
Photo by Tima Miroshnichenko
So I did something I had never done before. I opened an AI chatbot on my phone and typed out the whole messy question.
This post is about what happened next, what surprised me, and the rules I follow now. It is not medical advice. It is one parent being honest about a habit a lot of us are quietly picking up.
The Night I Asked a Chatbot Instead of Calling Someone
I want to be clear about why I reached for my phone. It was not because I trust a computer more than a doctor. It was because it was late, I was tired, and I did not want to wake anyone up over something that might be nothing.
That, it turns out, is exactly why most people do it. A KFF poll found that about a third of US adults (32%) had turned to AI chatbots for health information in the past year. Wanting a quick answer was the top reason.
The chatbot replied in seconds. It was calm. It was polite. It asked me my child’s weight, which honestly made me trust it a little more.
Then it gave me a dose. And this is where my real lesson started.
What AI in Healthcare Statistics Taught Me About Other Parents
The next morning, once everyone was fed and the fever had broken, I went looking. Was I the only one doing this? Was it a terrible idea?
I found a roundup of AI in healthcare statistics that made me feel a lot less alone. It reported that 46% of US adults have asked an AI tool a question about medication. It also found that most of those people worried the AI could give them wrong guidance, and yet many had still changed how they took a prescription because of it. You can read the full breakdown if you want the bigger picture on how patients and doctors are using these tools.
That combination stuck with me. We are using it. We are nervous about it. And we are acting on it anyway.
That is a pretty good description of how I felt at 11pm in my kitchen.
What the AI Actually Got Right
I do want to be fair here, because it was not all bad.
- It explained the label. It walked me through the difference between dosing by age and dosing by weight, and why weight is usually more accurate for kids.
- It reminded me about timing. It told me to check the minimum gap between doses on the packet and not to double up if I was unsure.
- It told me when to get help. It listed warning signs, like a very young baby with a fever or a child who is floppy and hard to wake, that mean you call a doctor straight away.
All of that was useful. It was the kind of thing a friend who happened to be a nurse might say.
Where It Fell Short
The problem was the number itself. When I compared the AI’s dose with the chart on the box, they did not quite match. The AI had rounded in a way that was close, but not the same.
Close is not good enough with children’s medicine. Their bodies are small, so small mistakes matter more.
It also gave the dose in teaspoons. The syringe in my drawer was marked in milliliters. Research shared by the NIH child health institute found that parents who thought of doses in teaspoons and tablespoons had twice the odds of making a dosing error. I had no idea until I looked it up.
And the chatbot did not know what it did not know. It had no idea what else my child had taken that day, or that the bottle in my hand was a different strength from the one it seemed to assume.
Why I Am Not the Only One Feeling Uneasy
When I told a few other parents at the school gate, the reactions were split. Some already used AI for everything from rash photos to sleep schedules. Others looked at me like I had said I let the dog do the shopping.
That split is real. A Pew Research survey found that 60% of Americans would feel uncomfortable if their own health care provider relied on AI to diagnose illness or recommend treatment.
I think both camps have a point. AI can be a brilliant explainer. It is a poor substitute for someone who can actually see your child.
My Five Rules Now
I have not deleted the app. But I use it very differently. These are the rules our house runs on now.
- The label wins. If the AI and the packet disagree, I follow the packet, or I ask a pharmacist. Every time.
- Milliliters only. I ask for doses in ml and I use the syringe that came with the medicine, never a kitchen spoon.
- I ask “why”, not just “how much”. AI is best at explaining. I use it to understand a label, then I check the number myself.
- Age limits are not suggestions. The FDA advises against giving over-the-counter cough and cold medicines to children under 2. If a chatbot ever suggests otherwise, I close it.
- When in doubt, call a human. A pharmacist, a nurse line or our doctor. That call is never a waste of anyone’s time.
What I Would Tell a Tired Parent at 11pm
If you are reading this in the dark with a warm little one on your shoulder, here is what I wish someone had told me.
It is fine to use AI to make sense of things. It is a decent study buddy. It can help you understand a confusing chart or think of the right questions to ask.
But it should never be the last word on how much medicine goes into your child. That job belongs to the label, your pharmacist and your doctor.
Write down the time of every dose. Keep the box, not just the bottle. And save your pharmacy’s number in your phone today, before you need it.
Final Thoughts
I learned more from that one night than I expected. Not really about medicine, but about myself.
I learned that I reach for the quickest answer when I am tired. I learned that a calm, confident reply can feel more trustworthy than it really is. And I learned that slowing down for two extra minutes to check the label is one of the kindest things I can do for my kids.
AI is going to be part of family life whether we like it or not. The trick is letting it help you think, without letting it do the thinking for you.