The Real Hiroshima Risk: Corporate Enclosure of Intelligence
Yvette Cooper warns of an AI catastrophe. She is right about the danger, but wrong about who is pulling the trigger.
Yvette Cooper’s recent essay for Chatham House warns that AI poses a danger on the scale of Hiroshima, and that we must act before we witness an “AI equivalent” of that catastrophe.
On the surface, the warning appears serious. In reality, it is a classic piece of Stage 4 Achiever misdirection.
The Foreign Secretary is right that AI carries profound risks. But she is wrong about where the greatest danger lies—and she is wrong about who should be trusted to manage it.
The Misdirection of Fear
The real existential threat is not rogue AI systems escaping government control. The real threat is the corporate enclosure of intelligence itself. Powerful technology companies, working hand-in-glove with governments, are rapidly concentrating the ability to shape thought, information, and decision-making into fewer and fewer hands. This is not a future risk. It is already happening.
When Yvette Cooper calls for stronger international rules on AI, she is not primarily talking about protecting ordinary people from corporate power. She is talking about managing AI in ways that preserve the existing structures of state and corporate authority. This is the same logic that has governed nuclear weapons, surveillance, and data: “safety” becomes a justification for centralised control.
Cooper’s record makes this framing especially troubling. She has long been a reliable supporter of Israeli government policy, including during periods of intense criticism over human rights abuses and military conduct in Gaza and the West Bank. Her willingness to align with powerful state interests while warning darkly about technological dangers reveals a consistent pattern: concern for control, not for liberation or genuine safety.
Reclaiming the Analogy
The Hiroshima analogy is revealing. The original Hiroshima was the result of a state weaponising advanced science for geopolitical dominance. The AI equivalent of that catastrophe is far more likely to come from the quiet, ongoing enclosure of intelligence by a handful of corporations and aligned governments than from some dramatic rogue superintelligence.
This enclosure looks like:
The Scraping of the Commons: Training data extracted without consent from billions of people.
The Optimization for Profit: Models engineered for engagement, behavioural prediction, and enclosure, rather than truth or human flourishing.
“Safety” as a Fence: Frameworks that conveniently protect corporate interests while marginalising open, sovereign, and decentralised alternatives.
The Replacement of Judgment: The steady displacement of human autonomy with opaque systems controlled by unaccountable entities.
The Path of the Commons
Yvette Cooper’s proposed solution—stronger government and international oversight—does nothing to challenge this trajectory. It simply seeks to ensure that the most powerful states and corporations remain the gatekeepers. This is not safety. This is managed enclosure.
The alternative is clear. We need Humble, Sovereign, and Decentralised AI—systems that are transparent, auditable, locally controllable, and designed to serve the living commons rather than extract from it. We need intelligence that remains rooted in human communities, not floating in corporate clouds or state surveillance architectures.
The real “Hiroshima moment” with AI will not be a single explosion. It will be the slow, quiet moment when most people no longer control the tools that shape their thinking, their information, and their choices.
That moment is approaching fast—and it is being enabled by the very institutions now warning us about danger.
We should not be fooled by dramatic language from those who have consistently chosen power over principle.
The task is not to hand AI to governments and corporations “for our own good.”
The task is to keep intelligence open, distributed, and accountable to the people it is supposed to serve.
That is the real work of safety. ✊🌎❤️