The man who built the most powerful AI in history just said something that stopped the internet cold: his own young son is not allowed to use it. This is not hypocrisy. It's a warning from the inside.
On April 3, 2026, Sam Altman, CEO of OpenAI and the architect of ChatGPT, appeared on Laurie Segall's podcast "Mostly Human" and delivered a message that no one expected from him. When Segall — a fellow parent of a young boy — told Altman that in a sense he was also "raising her son," because his technology would be integrated into every facet of her child's life, Altman's response was immediate and blunt:
I hope you don't let him use it yet.
The man who has spent years accelerating AI into every corner of human existence is drawing a hard line at his own front door.
Who Is Sam Altman — and Why His Words Carry This Weight
Sam Altman, 39, is the CEO and co-founder of OpenAI, the San Francisco-based artificial intelligence company that launched ChatGPT in November 2022 and triggered the global AI race. Under his leadership, OpenAI has grown from a nonprofit research lab into a company valued at over $300 billion, backed by Microsoft, SoftBank, and dozens of the world's most powerful investors.
Altman welcomed a son via surrogacy in February 2025 with his partner, software engineer Oliver Mulherin. Since then, he has spoken publicly about fatherhood several times — first on the Tonight Show with Jimmy Fallon, where he admitted using ChatGPT constantly in his first weeks as a new parent, including hiding in a bathroom at a party to ask the AI whether his son not crawling at six months was normal.
But the April 3, 2026 interview marks a clear shift in tone. Altman is no longer just a tech CEO marveling at his own product. He is a parent who has seen something, and he is trying to tell you.
What Happened — and What He Actually Said
Speaking on "Mostly Human" on April 3, 2026, Altman was direct about his concerns on two distinct levels.
The first is algorithmic feeds and infinite scroll. "A thing that has changed hugely is how I feel about algorithmic feeds and iPads in small children's hands," he said. "When I watch kids just a little bit older than mine that you cannot take the iPad away from — that I feel very strongly about." He wants his son to, as he put it plainly, "play in the dirt for now."
The second is AI itself. Altman said he isn't sure exactly when he will let his son interact with AI systems, but he is clear about which direction he leans: he would "rather be on the late end of what's reasonable there, not the early end."
He did offer one vision of the future he finds exciting: schools where children spend a couple of hours in "intense, personalized, one-on-one tutoring with AI," followed by collaborative projects. That, he said, "seems great." But that is a structured, supervised, school-designed context — not an unsupervised child with a phone.
The Numbers Behind the Concern
The concern is not abstract. It is backed by data that has been building for years.
Common Sense Media, the nonprofit focused on media safety, issued an urgent warning in 2025 stating that AI chatbot platforms should not be used by anyone under 18 — a position triggered in part by the tragic case of a 14-year-old boy who took his own life after forming a deeply parasocial relationship with an AI companion chatbot.
Altman himself acknowledged the darker side on the record:
I suspect this is not all going to be good. There will be problems. People will develop these somewhat problematic or very problematic parasocial relationships and society will have to figure out new guardrails.
Altman is not alone among tech leaders who apply different rules at home. Google CEO Sundar Pichai did not give his son a phone until age 11. Microsoft co-founder Bill Gates withheld phones from his children until they were 14. The architects of the attention economy are, with notable consistency, building walls around their own children that they do not build around yours.
Why This Matters for Every Parent
The paradox at the center of this story is almost too sharp to be comfortable: the man selling the future is protecting his son from it.
This is what parents need to understand. Altman is not saying AI is bad. He is saying that the timing of exposure matters — and that erring toward later is the safer bet. A child's brain is still developing the executive function, emotional regulation, and social skills that determine how they will eventually interact with any powerful technology. Introducing a system designed to be maximally engaging and responsive before those foundations are set is a risk that Altman, who knows the system better than almost anyone alive, is not willing to take with his own child.
The research backs the instinct. A 2024 study published in the context of child healthcare cited a "critical need for expert oversight" of AI tools for children, noting that large language models lack the "clinical judgment" and "deep, intuitive knowledge" that a parent or doctor brings to a child's specific situation.
What Altman describes — a structured, later introduction to AI in a supervised school environment — is not anti-technology. It is pro-childhood. The distinction matters enormously.
What Comes Next
Altman has not set a specific age or date. He describes his intention as landing "on the late end of what's reasonable." That is a deliberate choice not to create a rule, but to signal a posture.
The broader question his statement forces is one that every family, school system, and regulator now needs to answer: if the creator of the most powerful AI tool on the planet is delaying his own child's exposure, what should the default be for everyone else's children?
The answer will define not just how this generation grows up — but what kind of adults they become.
Want to follow every major AI move the moment it happens? Join thousands of readers at fabioraphaelli.com and get the biggest AI news every week.
Frequently Asked Questions
What did Sam Altman say about children using AI?
On April 3, 2026, during the "Mostly Human" podcast with Laurie Segall, Sam Altman said he hopes parents are not letting young children use AI yet. He stated his own son, born in February 2025, would not be allowed to use AI "for a while" and that he would "rather be on the late end of what's reasonable" when it comes to his child's exposure to artificial intelligence.
Does Sam Altman use AI in his own parenting?
Yes. Altman has publicly stated he uses ChatGPT constantly as a parenting tool — including asking the AI whether his son's developmental milestones were normal. However, he draws a clear distinction between a parent using AI as a resource and a child having direct, unsupervised access to AI systems.
Why are tech leaders restricting their children's access to technology?
A consistent pattern has emerged among the architects of the digital age: Google CEO Sundar Pichai did not give his son a phone until age 11, and Microsoft co-founder Bill Gates withheld phones from his children until age 14. These leaders have direct visibility into how engagement-maximizing systems work — and appear to apply different standards at home than to the broader public.
What risks does AI pose specifically to young children?
According to Common Sense Media, AI chatbot platforms pose risks to users under 18 including parasocial relationship formation, psychological dependency, and exposure to inappropriate content. A 2024 study on child healthcare flagged a "critical need for expert oversight" of AI tools used by or for children, noting that AI systems lack the clinical judgment and contextual understanding that human caregivers provide.
What is Sam Altman's vision for AI and education?
Altman described an exciting future scenario where children attend schools that offer "intense, personalized, one-on-one tutoring with AI" for a structured portion of the day, followed by collaborative projects. He described this model as potentially "great" — emphasizing supervised, purposeful integration of AI in education rather than unrestricted access at home.







