For the first time in the history of the U.S. intelligence community's annual threat assessment, artificial intelligence ranks above terrorism, cartels, and regional conflicts. Not as a footnote. As a priority threat. Davos, meanwhile, has stopped debating whether AI will transform the world and started debating how to survive that transformation. That shift in institutional language is the real news. Not the technology. The vocabulary.

When the institutions that define what counts as dangerous change their list of priorities, they aren't describing reality. They're building it.

The turn carries an irony hard to ignore. The same tools the U.S. developed to maintain global strategic advantage now appear in its own report as a vector of vulnerability. China has a declared plan to overtake it as the global leader in AI by 2030, and that plan isn't secret — they published it. The hunter is starting to notice that something larger is tracking it. This isn't new. The tools of dominance always find new owners. Gunpowder was a Chinese invention. So was the printing press. The internet was born in American military labs. Every transformative technology travels. And once it reaches different hands, whoever invented it rarely controls what happens next.

The militarization of AI isn't a future risk. It has already happened. Target-identification systems, real-time decision-making in conflict zones, predictive analysis of enemy movements. This isn't science fiction. It's operational infrastructure in conflicts happening right now. Technology designed with defensive purpose becomes an offensive tool before anyone can formulate a formal objection. That tendency doesn't require malice. It requires urgency, budget, and the absence of restraint. Those three ingredients are always available when there's a war on.

But what stands out most about the report isn't what it says. It's what it left out.

For years, climate change occupied the top spot in the threat reports of these same institutions. Existential threat, they said. Irreversible. Urgent. This year it vanished without explanation. No clarifying note. No transition. It simply stopped being there. The point isn't whether climate change is real or not — that's a separate discussion. The point is that a threat presented as the gravest in human history was withdrawn from the institutional narrative the moment the administration signing these documents changed. That confirms something researchers have documented for decades: threat narratives aren't built from data. They're built from political convenience. And whoever controls the threat narrative controls the budget, the legislative agenda, and public opinion.

This matters because it isn't the first time. The pattern holds consistent across decades: external enemy, internal threat, health crisis, climate collapse, technological hegemony. Each cycle generates its own structure of institutional response, its own apparatus of dominance, its own justification for concentrating power. The threat changes. The structure built around it stays.

The same report mentions Venezuela and Colombia. Not as actors with their own agendas, their own populations, their own history, their own legitimate interests. It mentions them as territories disputed by the U.S., China, Russia, and Iran over minerals and strategic positioning. Lithium, coltan, logistics routes. The language is that of a chessboard where the pieces have no names. From Mexico, these dynamics read differently. Not as abstract geopolitical analysis, but as a description of neighbors, of regions with a known history, of patterns that keep repeating on this continent. The same colonial scheme with new technology. The powers change. The process doesn't.

Neither Davos nor the CIA asks the question that matters most: why do systems designed to protect end up reproducing exactly the dynamics of control they claim to fight? It's not a rhetorical question. It's a design problem. When a system's stated goal is security, but its operational structure requires concentrating information, decision-making, and force, the outcome doesn't depend on the intentions of those operating it. It depends on the design. And the design, in every documented case, favors whoever already holds power.

There's something the historical record confirms with remarkable consistency: every time a transformative tool arrives in the world, the first question answered is never how it benefits the majority, but who controls it first. Cuneiform writing recorded debts before it recorded poems. The printing press was a tool of propaganda before it was a tool of emancipation. The internet was a military network before it was public space. AI doesn't break that pattern. It follows it at a speed we've never seen before.

Twelve thousand years of structures that promised security and delivered concentrated power. Not as a pessimistic argument. As a necessary diagnosis for understanding what kind of alternatives have any real chance of working. The historical record also shows that at the margins, in the cracks between dominant structures, there have existed forms of organization that distributed power differently. Not perfect. Not permanent. But real. Verifiable.

AI as a national security threat isn't the end of the story. It's the most recent chapter of a very long one. And the previous chapters suggest that the question isn't whether the tool is dangerous. The question is who decides how it's used, who watches the watchers, and whether we will ever design structures where that question has an answer that isn't always the same one.

Stones don't lie, but historians sometimes do.

Sources:

1. Annual Threat Assessment of the U.S. Intelligence Community — Office of the Director of National Intelligence (2023-2025 editions)

2. World Economic Forum Global Risks Report 2025 — World Economic Forum, Davos

3. China's Next Generation Artificial Intelligence Development Plan — State Council of the People's Republic of China, 2017

4. Autonomous Weapons and the Laws of War — International Committee of the Red Cross (ICRC), 2023-2024 reports

5. The Age of Surveillance Capitalism — Shoshana Zuboff, Harvard Business School Press, 2019