The Kingdom of Saudi Arabia has unveiled plans for the first city designed entirely by artificial intelligence. This is not a whimsical exercise in futurism. It is a stark announcement of a new epoch: one in which human intuition, messy democracy, and the organic sprawl of history are being systematically replaced by algorithmic efficiency. As I read the press release, I could not help but think of the Roman architect Vitruvius, who insisted that a city must be built on firmness, commodity, and delight. Our new masters have apparently decided that delight is optional.

The project, part of the Neom megacity initiative, promises a utopia of perfect traffic flow, optimised energy grids, and zero carbon emissions. It is, by all measurable standards, an improvement on the chaotic medieval alleyways of Jeddah or the traffic-snarled asphalt of Riyadh. But efficiency is a sterile god. It worships at the altar of the spreadsheet, not the soul. What happens to the crooked streets that inspired poets? What becomes of the accidental squares where revolutions were plotted? The AI cannot calculate serendipity.

Compare this to the Victorian era, when London's sewers were designed by Joseph Bazalgette with a vision that anticipated growth for decades. That was human intelligence informed by public health and civic pride. Today's AI is given a set of parameters: minimise commute times, maximise solar exposure, reduce water waste. It will produce a machine for living, but Le Corbusier already tried that, and we are still recovering from the brutalist towers he left behind.

There is also the question of national identity. Saudi Arabia is a young nation by historical standards, but it has a rich cultural heritage of mudbrick architecture, intricate geometric patterns, and the social fabric of the souk. Will an AI respect that? Or will it produce a generic glass-and-steel dystopia that could be transplanted to Dubai, Singapore, or Shenzhen without anyone noticing? The homogeneity of global architecture is already lamentable. This will accelerate it.

Of course, the proponents will argue that AI can be trained on historical data, that it can learn the nuances of local tradition. But AI does not understand meaning. It does not know why a mosque faces Mecca or why a courtyard is cooler in the afternoon. It optimises for outcome, not for significance. We are outsourcing our cities to a machine that has never felt the rain, never known the joy of a chance encounter on a street corner.

The Roman Empire fell, in part, because it became too complex to manage. Its roads, aqueducts, and bureaucracy were marvels of engineering, but they could not adapt to changing circumstances. They were brittle. Today, we are building a new Rome, one of sensors, algorithms, and predictive models. It will be even more brittle. What happens when the AI fails? Or when climate change makes its assumptions obsolete? We will have entire populations trapped in an infrastructure that cannot pivot.

Let us not mistake progress for wisdom. The Saudis may build their AI city, but they will find, as every empire has, that you cannot plan for the incalculable variable: human nature. We are messy, irrational, and wonderfully unpredictable. No algorithm can replicate that. And no city designed by one will ever be a home.