A Media Executive's Open Letter to Young People on AI Tells Them: It's Not Too Late

Imagine being handed a tool more powerful than anything your parents or grandparents ever touched, without an instruction manual, and being told the people who built it don't fully understand how it works. That's roughly the position young people find themselves in right now with artificial intelligence, according to a widely shared letter from Axios co-founder and CEO Jim VandeHei.

VandeHei first wrote to young people about AI in January. That column became, in his words, his most-shared piece ever. In the months since, he says, the technology has "gotten way better and at times scarier," prompting readers to ask what he'd say now, given the mounting warnings that AI could pose existential risks to humanity. His answer, published this week, is blunt: this generation has inherited a historic moment, whether it wanted it or not, and it cannot afford to look away.

To be clear, he's not describing a moment of war or economic collapse. The country, he notes, is "exceptionally peaceful and prosperous by historical measures." No one is starving. No one is fighting a civil war. But he argues something else is happening beneath that calm surface: we've built a technology so complex and fast-moving that even its creators can't say for certain what it will do next, and the risk, he writes, "rises by the hour."

Why AI Suddenly Feels Different

Most people's experience with AI is mundane. It's a smarter search engine. It drafts emails. Students use it to cut corners on homework. That everyday familiarity, VandeHei suggests, masks something much stranger happening underneath: the companies building these systems don't fully understand why their models behave the way they do, let alone what they're truly capable of.

Here's the plain-language version. Companies have poured enormous sums of money, mountains of data, and staggering amounts of computing power into building AI systems that aim to match or exceed human knowledge across every subject. The explicit goal, at many of these firms, isn't to slowly and carefully solve one problem at a time, like curing a specific disease. It's to create general superhuman intelligence and then point it at everything at once. Everything can mean curing cancer. It can also mean something far darker.

What has changed in the past six months, VandeHei argues, is speed. Timelines that once stretched a decade or more have compressed dramatically. Some in the industry now think true superhuman intelligence could arrive as soon as next year. Central to that shift is a concept called recursive self-improvement, sometimes shortened to RSI: the point at which AI systems, originally trained under human guidance, begin training themselves without human oversight. Think of it like a student who no longer needs a teacher, and starts writing their own textbooks. The catch is that we already don't fully understand how the human-guided version works. Layering self-improving AI on top of that uncertainty, VandeHei writes, is where things edge toward the "truly creepy and possibly catastrophic."

That kind of language, he admits, sounds hyperbolic. He acknowledges as much himself. But he says two years of deeply studying, using, and writing about AI has convinced him that young people aren't just living through history. They're going to help shape it, for better or worse.

The good news, as he frames it, is timing. We are still early. That means the outcome isn't fixed. Society has bent frightening, world-altering technologies toward good before, and there's no reason it can't happen again, provided people engage now rather than later.

He draws a direct line to earlier moments of American reinvention: the Civil War, the Great Depression, times when ordinary people did extraordinary things under extraordinary pressure. He calls this pattern "the genetic genius of America," the instinct to start big problems and then figure out how to solve them. His challenge to young readers is to see themselves as the next chapter of that same story.

The letter lays out concrete steps, though it resists calling them a formula. Learn about AI more seriously than any other subject in the news, not to obsess over worst-case scenarios, but because understanding the technology is quickly becoming a civic obligation. Separate hype from fear by asking who benefits from each claim: AI optimists promising cures and riches, critics warning of mass unemployment or extinction, and politicians who often understand the least but legislate the most. VandeHei argues the truth almost always sits somewhere between the hysteria and the hallelujah.

He also pushes back against a common impulse: boycotting AI out of principle. Refusing to touch the technology might feel like a stand, he writes, but it guarantees you'll never understand it well enough to shape it. His advice instead is to use it aggressively, to become the most AI-fluent person in your workplace, which he argues makes you both more valuable and more informed.

Underneath the practical advice sits an emotional one. Stay hopeful, but not passive. Pessimism, he writes, spreads like a disease. So does its opposite. He warns against slipping into "doomerism," the reflex to assume catastrophe is inevitable, while still taking the risks seriously enough to act.

Why It Matters

Perhaps the most human piece of advice has nothing to do with technology at all: touch grass, call a friend, watch a sunrise. VandeHei predicts that the more capable AI becomes, the more people will crave the things it can't replicate: love, sacrifice, joy, the feeling of standing on a mountaintop. Some of that wonder, he suggests, already got worn down by years of infinite scrolling, the pandemic, and political division. He sees this moment as a chance to reclaim it, not despite AI, but alongside it.

None of this comes with guarantees. VandeHei says he genuinely hopes the darker warnings prove overblown, that AI stays within human control and delivers on its promise to cure disease and expand prosperity rather than threaten it. But hope, in his telling, is not a strategy on its own. It has to be paired with attention, pressure on companies and lawmakers, and a willingness to actually use and understand the tools reshaping the workforce and society in real time. For the generation now coming of age alongside this technology, he argues, disengagement simply isn't an option anymore.