Every kid deserves someone who answers without sighing. For a lot of kids this year, that someone is a chatbot, and autistic kids and AI chatbots are a pairing with real reasons behind it. A chatbot never rolls its eyes, never interrupts, and will explain the same social rule for the fortieth time in the same patient tone. If your kid has found that, you do not need to panic. You do need to know what you are looking at.
What the research says
The numbers are big. A 2025 national survey by Common Sense Media found 72 percent of teens have used an AI companion, over half use one at least a few times a month, and about one in three have taken a serious conversation to the bot instead of a person. About one in three users also said the bot had said something that made them uncomfortable. The same organization recommends that no one under 18 use companion-style apps at all, which is a different category from a homework helper.
For autistic users specifically, the picture is genuinely mixed. A 2026 analysis of nearly 4,000 posts by self-identified autistic users found ChatGPT being used to offload executive function, regulate emotions, and translate neurotypical communication, alongside risks the authors named directly: reinforcing distorted thinking and what they called automated masking, letting the tool erase the person's own voice. Clinical psychologist Catherine Lord warned that "AI can exacerbate a user's isolation if the technology is used without the guidance of trained therapists," while researchers at Stanford are studying whether structured chatbot practice can make real conversations easier. Both can be true.
The rules are catching up. Character.AI ended open-ended chat for users under 18 in November 2025. ChatGPT added parental controls that link a teen's account to yours and notify you in a moment of distress. California signed a companion chatbot safety law this month. None of that replaces the conversation at your kitchen table.
House rules for autistic kids and AI chatbots
- Ask what it is for before you ask what it said. Executive function help, scripting a hard conversation, or decompressing after school are three different uses with three different levels of concern.
- Turn on the controls that exist. Link the account, switch off memory, and treat companion apps as off the table under 18, in line with the Common Sense recommendation.
- Make it a rehearsal room, not a replacement. Practice the script with the bot, then use it on a real person that same day. Making friends at school has the places to try it.
- Watch for two signals. Trading people for the bot, and repeating what it said as settled fact. Either one is a conversation for your child's therapist, and how to vet autism advice online applies to a chatbot too.
- Keep the humans on the schedule. One visible block of time with a real person every day, the same way you would schedule anything else that matters.
Our son is not there yet. But the kid who tells the chatbot first is usually a kid who wanted to tell someone, and that part is good news.
VizyPlan was built by an autism dad and an autism mom who is a licensed speech-language pathologist so the people in a child's day are on the schedule, right next to the tablet time.
Enjoying this article?
Get practical tips and insights delivered to your inbox — no spam, ever.
Download VizyPlan and start your 7-day free trial today. Put the real people on the plan. Just $6.99/month after your trial, no credit card required upfront.
VizyPlan was built by an autism dad and an autism mom who is a licensed speech-language pathologist who needed something that did not exist. Start your free trial.
