La OTAN advierte sobre el salto de los ciberataques al mundo físico gracias a la IA

Los ciberataques ya no se limitan al entorno digital. La inteligencia artificial está permitiendo que amenazas informáticas se traduzcan en impactos físicos reales, desde la interrupción de servicios esenciales hasta la operación de sistemas autónomos. En este contexto de desarrollo a pasos agigantados, organismos internacionales como la OTAN advierten de que este cambio marca un punto de inflexión en la seguridad global.

“Ya no hay una separación clara entre los ciberataques y los ataques físicos”, señaló la coronel Mietta Groeneveld, directora del Centro de Excelencia de Mando y Control de la OTAN, al referirse a la capacidad de la inteligencia artificial para operar sin intervención humana y trasladar su impacto al mundo real.

Industry experts agree that this scenario is already a reality: the rapid evolution of these technologies has caused their impact to no longer be limited to screens and to move into the real world.

AI industrializes cyberattacks

This change has a concrete technological basis. In recent months, models capable of detecting vulnerabilities in systems and autonomous agents that not only identify flaws have emerged, but can also act on them automatically. As a result, much of the attack process stops depending on human intervention. “for the first time, we are seeing technologies that not only analyze systems, but can decide and execute actions on their own”, explains Sergio Garcia, manager of i3e, Catalan technology company specializing in applied AI and cybersecurity.

Until now, exploiting a vulnerability required time and specialized knowledge. With artificial intelligence, that process is automated and drastically shortened. “An AI system can identify vulnerabilities that a team would take months to find and exploit them in a matter of hours”, García points out, who adds: “This completely changes the scale: we go from occasional attacks to continuous and automated attacks”.

This scenario multiplies the volume of threats and alters the traditional cybersecurity balance. 'The problem is no longer just sophistication, but the ability to launch massive attacks at minimal cost. for the first time, technology is favoring the attacker more than the defender'.

From cyberattack to physical impact

This change does not remain in the digital realm. When these capabilities are integrated into physical systems, the automation of the attack moves into the real world. “When these technologies are applied to devices or infrastructure, they cease to be a purely digital matter”, explains Garcia.

The development of autonomous drones illustrates this leap. These systems, supported by artificial intelligence, can identify targets, adapt their behavior and carry out actions independently, transferring the decision-making capability from the digital environment to the physical. 2It is at that point that NATO's warning is understood”, points out the expert, and develops: “Artificial intelligence is no longer just a tool, but an active component in conflict scenarios. When you automate the decision, automatizas también las consecuencias”.

Para i3e, este cambio marca un punto de inflexión: la inteligencia artificial deja de ser un riesgo tecnológico para convertirse en un elemento operativo con implicaciones directas en seguridad y defensa. “Estamos ante un cambio de paradigma que afecta a empresas, administraciones y, en última instancia, a la seguridad de todos”, García concludes.