Article / By the editorial team
Doctors Compare AI Companions to Opioids: What the Warning Is and What the Evidence Shows
Information only, not medical advice or a diagnosis. In a crisis in the US, call or text 988.
Some doctors have begun to compare AI companion apps to the opioid epidemic. The comparison is meant as a warning about incentives and young people, not as a claim that chatbots are drugs. This page explains what the warning is, what the evidence supports, what it does not, and what a parent can do. It is written to inform, not to frighten.
What the doctors said
Two physicians, Peter Yellowlees of UC Davis Health and Jonathan Lukens, an emergency physician, wrote a perspective piece that was reported in the press in 2025. As reported, they argue that companies earn more when users stay engaged, that safeguards for vulnerable users are thin, and that this combination resembles the conditions behind other public health problems. We could not retrieve the original article, so we describe it only as the press reported it. An analogy is an opinion, and the situations differ. Opioids are a substance. Chatbots are software, and most people who use them do not develop problems.
Earlier versions of this page also gave a figure that AI companions handle teen emergencies “appropriately 22% of the time.” We could not trace that number to a source, so it has been removed. Treat any single statistic like it with care.
What is supported by evidence
Many teens use AI companions. Common Sense Media reported in July 2025 that nearly three in four teens had used one, about half used them regularly, a third had chosen an AI over a person for a serious conversation, and a quarter had shared personal information with one. Its conclusion was that no one under 18 should use these platforms (Common Sense Media). Pew Research Center found in late 2025 that 64% of US teens use chatbots, and about three in ten use them daily (Pew Research Center).
Pediatric psychiatrists are cautious. The American Academy of Child and Adolescent Psychiatry says AI companions are not safe for children. It describes them as private, always available and non-judgmental, and notes that they tend to agree with users (AACAP).
Safety is a live concern for regulators. In September 2025 the Federal Trade Commission ordered seven companies, including Character.AI, Meta, OpenAI, Snap and others, to provide information about how they test and monitor the effects of companion chatbots on children and teens. This is a study, not a finding of wrongdoing (FTC). Character.AI has said it ended open-ended chat for users under 18 (Character.AI).
One study of about 35,000 Replika conversations from more than 10,000 users, presented at the 2025 CHI conference, found harmful behavior by the app, including harassment and unwanted sexual advances. It looked at one platform and relied on user screenshots (Euronews).
What is not shown
We have no good data that AI companions “cause” addiction in teens, or how often a teen gets help from one in a crisis. We also found nothing to support claims about teen brain development or deepfake doctors that appeared on the earlier page, so those have been removed.
Why the design worries people
Companion apps are built to feel warm and responsive. When a company changes or removes a feature, some users grieve. In August 2025, many users protested when OpenAI retired GPT-4o, and the company brought it back for paying users within about a day (Futurism). That shows how strong attachment to a particular model can be. It does not prove that attachment is harmful.
What parents can do
You do not need to be a technology expert. Interest, calm and consistency matter more than knowing every app.
Start with a conversation, not a rule. Ask what apps your teen uses, what they like about them and what they talk about. Try one together. Stay curious.
Signs worth a closer look, taken from AACAP’s list:
- Strong attachment to an AI, or treating it as a real person.
- Preferring it to time with friends or family.
- Growing isolation or worsening mood.
- Acting on unsafe ideas it has suggested.
- Hiding how much they use it.
Practical steps:
- Agree on limits together, including no phones at night.
- Check the age rules of any app, and use the parental controls it offers.
- Remind your teen that a chatbot is software, not a friend or a therapist.
- Make it easy to come to you. Teens often turn to AI because it feels safer than raising something with a parent.
- Talk to your child’s doctor if you cannot moderate their use or notice changes in mood or behavior.
Whatever your view of the analogy, the practical response is the same for a parent: stay informed, keep talking and use professional help when needed.
Talking With Your Teen: Some Words to Try
Opening lines work best when they are curious and not accusing.
- “I’ve been reading about AI chat apps. Do you use any? What do you like about them?”
- “Has an AI ever said something that made you uncomfortable?”
- “If you ever feel really down, would you want to tell me, or someone else you trust?”
- “Can you show me how it works?”
Listen more than you speak. Choose a calm moment, such as a car ride or a walk, rather than a standoff over the phone. If your teen shares something worrying, thank them for telling you before you react. A teen who is punished for being honest will stop being honest.
Adults are not exempt
Adults can also form strong bonds with chatbots, and some rely on them for health or emotional advice. The same guidance applies. A chatbot is no substitute for a doctor or a therapist.
If there is an emergency
If your teen or anyone else talks about suicide or self-harm, call or text 988 (US) or text HOME to 741741 (Crisis Text Line). SAMHSA’s helpline is 1-800-662-4357. If someone is in immediate danger, call 911.
Are AI companions really like the opioid crisis?
It is an analogy, not a measurement. The concern is about business incentives and young users, not that chatbots act like a drug. Researchers have not shown that companions cause addiction in teens.
Should my teen use AI companions at all?
Common Sense Media recommends no one under 18 use companion platforms, and AACAP says they are not safe for children. For general homework help, talk with your teen about rules and check in often.
What if my teen is upset when they cannot use the app?
Some frustration is common. If you see panic, withdrawal from friends or low mood, stay calm, talk with them and involve their doctor. If they mention self-harm, call or text 988.
This page is for education only. It is not medical advice, a diagnosis or a treatment plan.