By Stacy Bates, MS, RDN, SPAN Athletics
AI chatbots can be a helpful place to start a sports nutrition question, but they shouldn't have the final word on how a young athlete eats. Use them to learn terms and ask better questions. Then check the answer with a real person, ideally a registered or sports dietitian who knows your athlete.
Winter seasons are starting. Basketball, wrestling and swim athletes are asking what to eat to get stronger, faster or "bulk up," and many of them are asking a chatbot first. A Pew Research Center survey of U.S. teens, taken in fall 2025, found that 64% use AI chatbots and about three in ten use them every day. So the question isn't whether your athlete will use AI. It's how.
What the Research Says About AI and Sports Nutrition
Researchers have started testing how well chatbots answer sports nutrition questions. A 2025 study in PLOS ONE asked basic and advanced versions of ChatGPT, Gemini and Claude a series of sports nutrition questions. Experts scored the answers against the field's consensus statements. The results were all over the map. On open-ended questions, accuracy ranged from 31% to 74%, depending on the chatbot. On exam-style multiple-choice questions, scores ranged from 61% to 89%. The authors described chatbot accuracy as "moderate at best and inconsistent between models."
One more point for parents: those questions were about athletes in general. The study wasn't designed around kids and teens, whose needs change with growth.
Why Chatbots Sound So Sure
A chatbot doesn't know things the way a clinician does. It predicts the most likely next words based on what it has read. It sounds just as confident when the information is weak as when it is solid. Sometimes it makes things up, including studies that don't exist. The tech industry calls this a "hallucination." What these tools read matters too. A lot of the nutrition content online comes from influencers and brands, not experts. In a 2025 analysis of 250 nutrition posts on TikTok, dietitians created only about 5% of the content. When a chatbot learns from that kind of content, it can repeat the noise in polished, complete sentences.
That's a key idea in our book: confidence is not competence. A smooth answer isn't the same as a correct one.
What a Chatbot Can't See
When I sit with an athlete, I look at much more than a question typed in a box. I consider their growth, training schedule, sport, food preferences, what's available at home, and how they talk about food and their body. A chatbot sees none of that. It can't notice that an athlete has been more tired lately or is skipping breakfast. It can't follow up next week. If its answer is wrong, no one is accountable. That matters because a plan that sounds healthy can still leave a growing athlete under-fueled. Under-fueling over time can lead to relative energy deficiency in sport (RED-S), which affects energy, bone health, hormones and performance in both girls and boys. A real person in the loop is how you catch it early.
How to Use AI the Smart Way
USE IT TO LEARN, NOT DECIDE: AI is good at explaining a term like "glycogen," summarizing what a debate is about, or helping you write questions for your athlete's doctor or dietitian. That's a great use of a tool that's awake at 11 p.m. when no one else is.
ASK FOR SOURCES, THEN CHECK THEM: Ask the chatbot where its information came from. Then click through. Made-up answers often show up first in the sources.
RUN IT THROUGH THREE QUESTIONS: each your athlete to ask three things about any nutrition advice, from a chatbot, a creator or a teammate:
Who is saying this?
What are they selling?
What does the rest of the evidence say?
Treat absolute certainty and numbers as a yellow flag. If a chatbot gives your child a precise calorie, gram or "per pound" target, pause. Needs for growing athletes depend on age, growth stage, sport and training load. A number like that belongs in a conversation with a pediatrician or registered dietitian.
It’s also important to know the RED Flags. Be cautious with any answer that tells a young athlete to cut out whole food groups, eat less, focus on weight, or add supplements, pre-workouts or energy drinks.
What I See as a Dietitian and a Mom
I use these tools myself. They help me find a study I half-remember or draft a handout faster. But after nearly 20 years as a registered dietitian, and as a mom of two athletes, I also know how quickly a confident answer can send a family in the wrong direction. One of my boys showed me a chatbot answer recently about which pre-workout to take. It listed ingredients and doses with total confidence. It never asked how old he was. We talked about why caffeine, sleep and a growing body don’t mix well, and what he could eat before a game instead. He wasn’t trying to cut corners. He just asked the tool that was in his pocket. Another recent example, an athlete I work with came to a session with a meal plan a chatbot had built her. It looked polished, and it was about 800 calories short of what her training called for. We went through it line by line, and the conversation we had afterward was the real value: how to tell when an answer fits you and when it only sounds like it does.
You don't need to ban AI. Ask your athlete to show you what they asked and what it said. Read it together. Talk through the three questions. Kids who learn to think critically about what they read build a skill that helps them on and off the field.
Keep a Real Person in the Loop!
SPAN Athletics and Nutrition for Performance work with athletes, families, coaches and teams on practical, evidence-based fueling and mindset. If your athlete has questions a chatbot can't answer well, reach out to us. For more on cutting through online nutrition noise, check out our book, co-authored with Tony Castillo, RDN, Unfollow the Noise.