Colorado Poll Shows 63% Fear AI Will Destroy Humanity

The cursor blinks on the screen, a rhythmic pulse in the quiet of a home office. Outside, the September wind ruffles the leaves of the cottonwoods lining the street, but inside, the tension is digital. It’s the kind of silence that follows a question no one wants to answer: Are we actually in danger?

That is the question locals are asking as the artificial intelligence debate moves from Silicon Valley boardrooms to our own living rooms. It’s not just about whether your job will be automated next year. It’s about whether the technology itself might outgrow us entirely.

The fear is no longer theoretical. According to a recent Politico poll, 63% of Americans now believe there is a moderate to near-certain possibility that AI will destroy humanity. That’s a staggering shift from simple concern to existential dread. And it’s not just the general public feeling the jitters. Actual AI researchers are warning that there is more than a 10% chance we could all be killed by a self-regulating, self-teaching, super-intelligent AI by the end of the decade.

Ten percent is not nothing. In engineering and finance, a one-in-ten chance far exceeds standard safety thresholds. It requires active mitigation, not dismissal. Yet, as Mike Littwin points out in the Colorado Sun, we are being told by some industry leaders — and notably by President Trump — that we shouldn’t worry. Trump has characterized the AI dangers as largely hype, suggesting that the "high IQ" crowd is overreacting.

But the skepticism runs deep. Why do we trust the executives selling the software to be the best judges of its safety? As Littwin notes, we know better than to put a huge amount of trust in AI, or in industry executives who may be asking for complicated regulation to keep control of AI for themselves. The irony is palpable: the very people building the engine are telling us the fire isn’t hot.

The stakes are raised further by international players. Reports indicate that China and Iran are already using AI in advanced ways to spread political propaganda. If our democracy is vulnerable to algorithmic manipulation, the threat is immediate and tangible. It’s not just about a rogue AI agent convincing your doctor to perform unnecessary brain surgery; it’s about the erosion of shared reality.

So, what do we do? The call to action is coming from an unlikely coalition. Nearly everyone is demanding government intervention, from Bernie Sanders to Steve Bannon, from Elon Musk to Sheryl Crow. The goal is to get the government to put in regulations that don’t kill the science but also don’t kill us. We need companies to be transparent about the gains and the dangers. We need cooperation with China, because if humanity is truly on the line, national borders matter less.

The work starts now. It’s messy, complex, and urgent. But standing still is no longer an option. The covers won’t help. We have to look at the machine, understand its gears, and decide if it’s time to hit the brakes.

The cursor still blinks. The wind outside picks up, rattling the windowpane. The question hangs in the air, waiting for an answer that feels further away than ever.