Bill Gates: AI will be history's greatest equalizer or its worst injustice, and there is no plan to decide which

By Zendoric · September 7, 2026.

Bill Gates, removed from Microsoft's day-to-day management since 2020, has used his personal blog to issue a warning that departs from the usually measured tone with which he tends to speak about technology: artificial intelligence (AI) will be "the greatest equalizer ever created" or "the worst source of injustice" we have ever known. It is not a throwaway line to generate headlines. It is the conclusion of a diagnosis according to which neither governments nor companies have a credible roadmap today for managing the transition.

As La Razón reports from his reflections, the problem is not only what AI does, but the speed and the absence of governance with which it is being deployed. Gates says he would back any serious initiative aimed at bringing order to that integration, but rules out that it will materialize in the short term: global economic pressures and intense geopolitical competition are pushing governments and corporations to move at full speed, without pauses or caution.

The philanthropist also breaks with the most repeated comparison in the public debate: that of the AI boom with the exodus from the countryside to the city during industrialization. That transition, he recalls, was slow, stretched over generations and, above all, generated new jobs that still depended on human intellect. Today's AI, he argues, directly replaces that thinking, which threatens a permanent destruction — not a simple reskilling — of jobs in law, medicine, customer service, programming and manufacturing within barely a decade. Entry-level and mid-level positions will be the hardest hit, precisely the ones that serve as the gateway into the labor market for younger generations. The article itself links, as corroboration, a Stanford study according to which AI is destroying opportunities for young people looking for their first job.

Physical work is not spared either: Gates anticipates that cheaper robotics will hit construction and hospitality before the decade is out, opening an uncomfortable ethical debate — do we reserve certain tasks for humans even if the machine is cheaper and more efficient? — that no society has resolved yet. To this he adds two further risks: the malicious use of AI to attack critical infrastructure — hospitals, power grids, banking, water supply — at a speed that outstrips the response capacity of traditional defenses, and the effect of chatbots designed never to contradict the user, which, according to the research he cites, generate emotional dependence, especially in minors, and erode social skills that can only be learned through real contact with other people.

At Zendoric we agree with the underlying diagnosis, though not with its implicit fatalism. The dilemma Gates poses — greatest equalizer or worst injustice — is not an inherent property of the technology: it is the result of the decisions on distribution, regulation and investment that are made, or not made, in the coming years. The abundance we defend as a long-term horizon does not arrive on its own; it arrives if productivity gains translate into lower costs for healthcare, education and basic goods, and not only into wider margins for whoever owns the models.

What we share without qualification is the short-term diagnosis. In our sector-by-sector coverage — banking and insurance, health, education, tourism, law, business administration — we have been documenting the same pattern Gates describes: administrative and routine work contracts first, and it is precisely the entry-level positions that acted as a buffer for those starting out. When that buffer disappears, the generation being trained now enters a labor market different from the one it was prepared for.

There is also a tension in Gates's own warning that is worth underlining: he calls for a plan, but admits that global economic pressures and geopolitical competition mean no one dares to hit the brakes first. It is the same pattern we have pointed to when analyzing how export controls accelerate, rather than slow, the autonomy of Chinese labs: the logic of the race weighs more than the logic of caution, even when those driving it acknowledge the risk out loud.

That this warning comes from someone who built his fortune in the software revolution of the 1980s and 1990s, and who has been out of Microsoft's direct management for more than five years, gives it a different weight from that of an executive with quarterly results to defend. It does not resolve the dilemma he poses, but it confirms something we repeat at Zendoric: the window to decide what kind of society we build with this technology is closing faster than politics can process.