On Tuesday, the FBI announced that an OpenAI engineer had been arrested for allegedly bugging an executive conference room at the frontier lab’s San Francisco offices. The device, a $14 Chinese-made development board, had been wrapped in toilet paper and plugged into the back of a television, and the suspect had left a bag in another room containing his ID and a second recording device. The would-be Watergate plumber had apparently used a chatbot account linked to his wife to ask how to build a continuous audio recorder. It was a bold move, given that the chatbot belonged to the company he was trying to bug.

Questions abound, chief among them who the Bay Bugger is and why he would jeopardise such a lucrative career. Abdulrahman Alfozan, who joined OpenAI in July 2025, was the company’s only Saudi employee. His LinkedIn profile also lists him as a venture partner at RAED Ventures, a Saudi firm where, according to the investigative journalist Scott Stedman, he sits alongside the son of a senior Saudi intelligence official.

As for his motive, nobody knows yet. He has been charged only with illegally intercepting oral communications, an offence carrying a sentence of up to five years, and prosecutors have yet to say what he intended to do with any recordings. Whatever he was after, it would have been worth far more than any pay packet: a single conversation in a frontier lab’s boardroom could save a rival years of research and billions of dollars, while buying it a head start in a future that is still all to play for.

That is the measure of what AI has become: the defining technology of our age, redrawing the world as radically as the steam engine, the bomber and the atom did before it. And just as the nuclear age divided the world into those with the bomb and those without it, the AI age is splitting it into those who build the technology and those who merely consume it.

The AI-maker in chief is the United States, which holds the leading frontier labs, the most capable models and the designs for the chips on which they run. This makes it the obvious target for anyone hoping to claw their way ahead. From the Gulf to Europe to India, every government now treats AI as central to its national agenda, and Saudi Arabia, which is betting heavily on AI to wean its economy off oil, is no exception.

Washington understands this, which is why frontier AI now carries the same strategic weight as the atomic bomb once did — a comparison the labs have been happy to encourage. Little wonder the Justice Department described the theft of superintelligence technology as a serious threat to national security, just a fortnight after the Director of National Intelligence, Jay Clayton, made the same case for superintelligence itself.

The paranoia is not unfounded. In January, a former Google engineer became the first person convicted of AI-related economic espionage in the US. Meanwhile, American labs have accused Chinese rivals of copying their models and even smuggling. In October, a California executive was charged with routing more than $300 million in Nvidia-equipped servers to China via Malaysia and Singapore.

There is no question that the stakes are high. Donald Trump said last month that “whoever wins AI, wins” — and he’s right. A lead in AI compounds into leads in science, industry, intelligence and war, each of which feeds the next round of AI. There is no real middle ground.

Getting a job at a frontier lab may soon look like getting one in national security. It will require clearances, background checks and scrutiny over foreign ties — a painful adjustment for an industry built on hiring the best from anywhere. Espionage is the price of the frontier, and there will no doubt be more industrial bugging to come.