AI Will Be Met with Violence, and Nothing Good Will Come of It

What was published

Alberto Romero published a polemic on Substack arguing that the social response to advanced AI will not be code or regulation, but violence. The piece — which has been circulated on Hacker News — traces a line from fragile looms and 19th‑century Luddites to modern, hardened datacenters, and concludes that when machines become too robust to attack directly, people will become the obvious weak link. It has been reported that the essay cites a recent episode in which Iran’s Revolutionary Guard released satellite footage of OpenAI’s Stargate campus in Abu Dhabi and allegedly threatened “complete and utter annihilation.”

Luddite echoes, modern stakes

Romero leans into a vivid historical parallel: when technology outpaces social control, rhetoric hardens and hands reach for guns. He recalls Luddite-era violence and the assassination of a mill owner as an emotional fulcrum — a reminder that anger at industrial change has a human, sometimes bloody, face. The image is stark: break the loom, the factory, the machine — or, if you can’t, break the people who keep them running. Punchy, unsettling, and hard to shake.

Why this matters now

Whether you call the essay alarmist or prescient, it lands at a fraught intersection of tech, security, and politics. The piece reframes the threat model: not only do we need sturdier infrastructure and better governance, we also need to reckon with social backlash that could target workers, researchers, or whole communities. That’s a policy problem as much as a technical one. If nothing else, Romero’s argument is a provocation: can we design AI futures that don’t turn people into the soft underbelly of hardened machines? It’s a question worth asking — loudly.

Sources: thealgorithmicbridge.com, Hacker News