August 2026 became a turning point in the understanding of artificial intelligence. Three key figures—Mark Zuckerberg, Bill Gates, and writer Cory Doctorow—presented the world with diametrically opposed visions of a future where AI rules the roost. However, they are all united by one troubling detail: confidence in the effectiveness of a technology that, upon closer inspection, proves to be shaky.
A rift in the vision of the future
In his manifesto, Zuckerberg paints an idyllic picture: personal superintelligence for every inhabitant of the planet, working for the benefit of humans in a fully private mode. The head of Meta, whose market capitalization reaches $1.47 trillion, promises a return to open source and the creation of an auction mechanism for distributing computing power. Yet behind this progressive rhetoric lies a fundamental question: will a person gain a tool or become a cog in it?
Gates, by contrast, sounds the alarm. He calls the upcoming transition "one of the most turbulent periods in history" and honestly admits: there is no plan for entering the new era. His proposals—from a "reserve" for professions to a tax on AI tokens—sound reasonable but run up against the absence of institutions capable of implementing them.
The harsh reality of consulting
The most sobering was the report by Nikhil Suresh from Hermit Tech. After conducting about 300 interviews with professionals around the world, he uncovered a frightening pattern: with a zero share of successful AI projects, corporations continue to report a hundredfold increase in productivity. Top managers making strategic decisions often have never opened ChatGPT, while employees inflate "token leaderboards," creating meaningless duplicate projects.
The numbers speak for themselves: even a perfectly tuned Cortex add-on makes errors in 8% of cases, and someone must catch each error. The paradox is that verifying a machine's result costs almost as much as producing it—it's just that no one pays for those hours.
The "reverse centaur" phenomenon
In his book, Doctorow flips the classic "centaur" concept, where the human is the head and the machine is the body. In the new reality, the algorithm makes decisions, and the human performs physical work at an "inhuman pace." Drivers under recognition cameras, programmers checking someone else's code, lawyers signing documents on behalf of AI—they all become a temporary measure until the frequency of errors justifies their salary.
Data from the Stanford Digital Economy Lab confirms: employment of young people aged 22–25 in professions highly exposed to AI is 19% lower than that of their peers in less affected fields. The gap is growing, and this is not about layoffs—it is the quiet disappearance of hiring.
My verdict
While billionaires argue over the distribution of access to technology, a more fundamental question is being decided—who will submit to it. The market is already giving the answer: companies are cutting staff under the banner of AI, even without measurable results, and investors punish stocks at the slightest doubt about returns. In this system, the human is merely a temporary controller whose place will vanish the moment algorithms achieve a "virtually error-free" result. The question is not whether that moment will come, but whether we will have time to create institutions capable of protecting human dignity before machines become truly indispensable.