Did the genie escape from the bottle?

By: Beatrice E. Rangel - 15/09/2026

September promises to be a month of surprises. Adding to the puzzle that the selection of the next UN Secretary-General has become are the plans of Northern Ireland, Scotland, and Wales to organize their independence from the United Kingdom. And before we've even recovered from this news, Dario Amodei, president of Anthropic, one of the leading firms in the development of artificial intelligence, published an article in which he calls on the industry to slow the pace at which this technology is advancing.

According to Amodei, like many previous technologies, artificial intelligence carries risks. But precisely because it is such a powerful technology, these risks are particularly serious. Among them, he mentions the possibility of losing control of AI systems, their use for cyberattacks and bioterrorism, and their capacity to cause severe economic disruption. He also warns that unbridled competition to offer increasingly more artificial intelligence services at the lowest price could exacerbate these dangers.

Interestingly, back in 2018, The Atlantic magazine had already published an essay by Henry Kissinger entitled "The End of the Enlightenment," in which he warned about the potentially destructive capacity of artificial intelligence and even compared it, due to its implications, to nuclear weapons.

For Kissinger, humanity was neither philosophically nor intellectually prepared to manage artificial intelligence. His concern went beyond the technology itself: complex human decision-making systems cannot be reduced simply to data. AI, he warned, develops driven by its own logic and can produce conclusions that humans are unable to fully explain or reconstruct, thus challenging traditional forms of cognition and even our historical concepts of reality.

The warnings of Henry Kissinger, also supported by Eric Schmidt, former president of Google, have not been the subject of a public debate proportionate to the magnitude of the challenge, nor do they seem to have received sufficiently serious consideration from the world's leaders.

The reason is perhaps simple: almost all of them are caught in a whirlwind of the present that prevents them from developing a forward-looking and comprehensive vision.

Governments grapple daily with economies that have lost dynamism and fail to generate enough jobs to guarantee a stable life for young people entering the workforce. Transnational organized crime has infiltrated numerous citizen protection systems, while cybercrime proliferates at a rate that authorities can barely contain. Higher education, which for decades served as a bridge from poverty to the middle class, is creating a generation of young professionals in the United States burdened by student debt that early 21st-century salaries can hardly afford to repay.

All these pressures force governments to focus on day-to-day affairs. Meanwhile, in technology labs, innovation continues at full speed.

Hence, relatively little thought has been given to building the necessary channels to channel the extraordinary creative energy of artificial intelligence towards enriching the activities that make up the network of services in modern society; to establishing new educational paradigms capable of preparing future generations to live and work with it; or, as Kissinger argued from a geopolitical perspective, to developing deterrent mechanisms capable of preventing this extraordinary creative capacity from also becoming a capacity for destruction.

To further complicate matters, artificial intelligence has become the spearhead of economic and strategic competition between the two major powers vying for global leadership: China and the United States.

China has made progress in integrating artificial intelligence into its production systems, healthcare systems, and, still in its early stages, its education networks. In the United States, however, the debate continues about what to do with it, how to regulate it, and to what extent to allow it to advance without restrictions.

And if the prevailing conviction in Washington is that any significant regulation could translate into a loss of ground to China, it is difficult to imagine that, in the midst of this race, a sufficiently deep public reflection can emerge on how to channel one of humanity's most extraordinary creations towards increasing well-being, strengthening societies, and fostering international understanding.

That, precisely, is the danger.

We don't yet know if the genie has escaped the bottle for good. But we do know that we've started chasing after him without first deciding where we want him to lead us.

And if an accident caused by an out-of-control artificial intelligence ever occurs, we may discover too late that its destructive capacity could surpass that of Chernobyl. The difference is that Chernobyl was an accident at a specific facility. Artificial intelligence, on the other hand, knows no boundaries.

«The opinions published herein are the sole responsibility of its author».