Public narratives about science are often shaped less by data than by incentives. When storytelling replaces evidence, we risk stifling innovation that could solve real problems — and ignoring the need for sensible safeguards where they’re actually warranted. Both outcomes endanger public safety and erode trust in science.
Nowhere is this clearer than in the unfolding convergence of nuclear energy and artificial intelligence.
The failure to scale nuclear power remains one of the great moral and strategic tragedies of the modern era. Over 8 million lives are lost each year to fossil fuel-related pollution. Billions live without access to reliable energy, stunting economic growth, hindering industry and deepening poverty. Forests are cleared for agriculture where nuclear-powered greenhouse farming could have fed millions. Freshwater shortages, geopolitical instability and dependence on hostile oil regimes all trace back to one failure: We abandoned the promise of nuclear energy, not because the science demanded it but because incentives aligned against it.
The safety profile of nuclear energy has been well established. The International Atomic Energy Agency reports that nuclear causes fewer deaths per terawatt-hour than oil, wind or hydro. Oil results in 18.4 deaths per terawatt-hour, while nuclear accounts for only 0.03. Even factoring in high-profile accidents like Chernobyl and Fukushima, nuclear power remains remarkably safe. Modern reactor designs have only improved these margins.
So why did we turn our backs on nuclear? Not because the science changed, but because the story did — and because too many actors benefited from telling the wrong one. Misinformation, media sensationalism, fossil fuel lobbying and activist fear campaigns combined to create a stigma and opposition that persists even among climate advocates.
Thankfully the tide in public discourse on nuclear may finally be turning. After decades of fear-mongering, even critics are starting to acknowledge what the data has always shown: nuclear is among the safest, cleanest and most scalable sources of energy available.
While the vibe shift on nuclear is promising, the catalyst should give us pause. The change is led by a powerful sector now having skin in the game. The tech industry seeks to build nuclear reactors to power data centers needed to realize their aspirations of creating artificial general intelligence.
The very companies demanding nuclear reform are also fast-tracking artificial general intelligence, a powerful, high risk and poorly understood technology with little to no regulatory oversight. There are no mandated safeguards against catastrophic failure, liability frameworks, kill switches or red lines for deployment. A group of leading AI scientists recently noted, “There are fewer regulations on AI systems that could pose catastrophic risks than on sandwich shops or hairdressers.”
This is the reverse of the nuclear problem. Where nuclear was paralyzed by fear despite clear safety data, artificial general intelligence with well documented risks is being rushed forward under a techno-optimist banner without serious scrutiny. Once again, incentives trump science.
This pattern of compelling storytelling reigning supreme over clear science is not limited to nuclear and AI. Genetically modified crops have passed rigorous safety testing and offer enormous potential to feed the world sustainably. But European bans — driven by fear, not fact — block their adoption. Conversely, many food additives that have been shown to cause health problems are rightly banned in Europe but remain legal in the United States, much to the convenience of Big Food. As a government report on biotechnology noted, “Regulation is often written to combat perceived risk rather than actual risk.”
If we want to unlock an abundant, secure and innovative future, we must stop regulating based on vibes and start regulating based on verifiable science. That means defending innovation when the evidence supports it — like with nuclear and genetically modified organisms. We would have to defend real accountability when there are risks needing mitigation, such as artificial general intelligence and food additives.
The public shouldn’t have to choose between reckless acceleration and anti-tech stagnation. There’s a third path: progress with safeguards. But those safeguards must be designed by independent scientists and new institutions aligned with truth seeking — not corporate lobbyists, not public relations teams, not the same actors who failed to question, or even amplified, convenient false narratives the first time around.
The opportunity before us is enormous: energy abundance, disease cured and sustainable growth for billions. But if we repeat the mistakes of the past, we risk missing that future — or worse, creating one we cannot control. It’s time to flip the script and realign incentives to hold every narrative, whether utopian or dystopian, to the standard of truth.
Emilia Javorsky is director of the Futures Program at the Future of Life Institute.
Copyright 2025 Nexstar Media, Inc. All rights reserved. This material may not be published, broadcast, rewritten, or redistributed.
For the latest news, weather, sports, and streaming video, head to The Hill.
