In the future, it will never move this slowly again
En Bletchley Park, cuna histórica del descifrado moderno, el Secretario General de la ONU António Guterres advirtió esta semana que la humanidad se encuentra en un umbral sin precedentes: los sistemas de inteligencia artificial avanzan a una velocidad que ningún marco de gobernanza existente ha logrado acompañar. Su llamado no fue a detener el progreso, sino a recordar que las herramientas más poderosas de nuestra era podrían profundizar las desigualdades más antiguas del mundo si no se guían por principios compartidos. La pregunta que dejó suspendida en el aire no es si la IA cambiará el mundo, sino si el mundo tendrá algo que decir sobre cómo lo hace.
- Los modelos de IA más potentes se despliegan globalmente con mínimas salvaguardas, abriendo la puerta a su uso por parte de criminales, terroristas y actores que buscan corromper la información.
- La brecha digital amenaza con convertirse en un abismo permanente: 21 de los 25 países menos preparados para la era de la IA están en África, quedando excluidos de sus beneficios económicos y sociales.
- Guterres advirtió que la tecnología no solo no frenará su ritmo, sino que acelerará indefinidamente, haciendo que cada momento de inacción sea más costoso que el anterior.
- La comunidad internacional intenta construir consenso a través de una serie de cumbres —Reino Unido, Corea del Sur, Francia— pero la carrera entre la gobernanza y el desarrollo tecnológico aún no tiene ganador claro.
- El llamado de la ONU no propone reinventar los principios éticos, sino anclar la regulación de la IA en marcos ya existentes: la Carta de la ONU y la Declaración Universal de Derechos Humanos.
António Guterres tomó la palabra esta semana en Bletchley Park, Inglaterra, en una cumbre dedicada íntegramente a la seguridad de la inteligencia artificial. Su mensaje fue directo: la velocidad a la que avanza esta tecnología no tiene precedente histórico, y lo más inquietante es que nunca volverá a ser tan lenta como hoy.
Durante el último año, modelos de IA de gran potencia han sido lanzados al mundo con escasa atención a medidas de seguridad. Guterres advirtió que cada despliegue irresponsable multiplica los riesgos: estos sistemas pueden ser utilizados como armas por criminales o terroristas, pueden distorsionar la información y pueden evolucionar en direcciones que nadie anticipó.
Pero la crisis más inmediata, señaló, no es la de las máquinas fuera de control, sino la de la desigualdad. Ningún país africano figura entre los cincuenta más preparados para la era de la IA, y 21 de las 25 naciones menos listas pertenecen a ese continente. La misma tecnología que podría mejorar la salud, la educación y la economía de millones de personas amenaza con ensanchar las brechas que ya dividen al mundo.
Guterres no pidió frenar el desarrollo de la IA, sino gobernarlo con urgencia y desde principios ya establecidos: la Carta de la ONU y la Declaración Universal de Derechos Humanos. La cumbre de Bletchley Park es apenas el inicio de un proceso que continuará con encuentros en Corea del Sur y Francia. La pregunta abierta es si el consenso internacional podrá construirse antes de que la tecnología supere por completo la capacidad humana de orientarla.
António Guterres stood at Bletchley Park in central England this week and delivered a warning that has become harder to dismiss with each passing month: the systems we are building to think for us may soon think in ways we cannot predict or control.
The UN Secretary-General was speaking at a summit devoted entirely to artificial intelligence safety, and his message was blunt. The speed at which AI technology is advancing has no historical precedent. More unsettling still, he suggested, is what comes next. "The paradox is that in the future, it will never move this slowly again," Guterres said. He was not being poetic. He meant it as a statement of fact.
Over the past year, powerful AI models have been released into the world with minimal attention to safety measures or user protections. Each time this happens, Guterres argued, the risk multiplies. These systems could be weaponized by criminals or terrorists. They could corrupt the integrity of information itself. They could develop in directions no one anticipated or intended. The concern is not hypothetical—it is baked into the current trajectory of development.
But the immediate crisis, Guterres suggested, is not about runaway machines. It is about inequality. No African nation ranks in the top fifty countries for AI readiness. Of the twenty-five nations least prepared for the AI era, twenty-one are in Africa. This is not a future problem. It is happening now, as some parts of the world race ahead while others are left further behind. The technology that could help governments budget more efficiently, help businesses grow, help climate scientists predict droughts and storms, and help ordinary people access healthcare and education—this same technology threatens to widen the chasms that already divide the global economy.
Guterres did not argue for halting AI development. Instead, he called for "new solutions" built on foundations that are not new at all: the UN Charter and the Universal Declaration of Human Rights. Governance frameworks for artificial intelligence, he suggested, should be rooted in principles that have already been established, not invented from scratch. The world needs immediate action to prevent AI from making existing inequalities worse. It also needs sustained, unified, global coordination—the kind of multilateral response that has become increasingly rare.
The summit at Bletchley Park is being treated as a beginning rather than a conclusion. South Korea will host a virtual follow-up in six months. France will host another in-person gathering the following year. Guterres called this progression "an important step on the road toward consensus." What remains unclear is whether consensus can be built fast enough, or whether the technology will simply outpace the conversation about how to govern it.
Citações Notáveis
The speed and scope of current AI technology is unprecedented. The paradox is that in the future, it will never move this slowly again.— António Guterres, UN Secretary-General
We have seen the launch of powerful AI models with little consideration for safety and protections for users. Each time this happens, it increases the risk that the technology will be used maliciously by criminals or terrorists, or that it will develop in unforeseen directions.— António Guterres
A Conversa do Hearth Outra perspectiva sobre a história
When Guterres says AI could spiral beyond human control, is he talking about something like the machines turning against us, or something else?
He's describing something more immediate and less dramatic. He means that as these systems become more complex, we may not be able to predict what they'll do or why they do it. A criminal could use one to forge documents. A terrorist could use one to plan attacks. The system itself doesn't need to be conscious or rebellious—it just needs to be powerful and opaque.
So the real danger is that we've built something we don't fully understand, and we're deploying it anyway?
Exactly. And we're doing it at speed. He emphasized that point repeatedly—the pace is unprecedented. There's no historical comparison for how fast this is moving.
But he also talked about Africa being left behind. How does that connect to the control problem?
It's a different kind of risk, but equally serious. If AI development is concentrated in wealthy nations, and African countries can't participate or prepare, then the technology that could help them will instead be imposed on them. They'll be shaped by systems they didn't build and can't govern.
Is there a way to slow things down, or is that impossible?
Guterres didn't call for slowing down. He called for governance frameworks—rules, oversight, international coordination. Whether those can be built fast enough is the real question. The summits in South Korea and France are meant to start answering it.