With the emergence of agent-based systems, a new chapter in the digital age is beginning.

Systems no longer merely execute instructions: they perceive, reason, plan, and act. They interact with one another, make decisions, and alter their environment. Artificial intelligence is becoming a full-fledged player.

This transition is causing a sense of vertigo.

First, a sense of technological vertigo, as autonomous capabilities operate at an unprecedented speed and scale. Then, a sense of organizational vertigo, as decision-making chains fragment among designers, operators, and users. Finally, a sense of strategic vertigo, as we must constantly strike a balance between performance, security, and accountability.

In this new landscape, the role of humans is undergoing a profound transformation. Humans no longer validate every action in real time. They define rules, set limits, and oversee systems already in operation. They govern decision-making loops that they no longer directly control.

In this context, cybersecurity itself must evolve.

The challenge is no longer limited to protecting systems or safeguarding data. It involves securing artificial entities capable of taking initiative and acting. These come with new risks: hijacking of objectives, manipulation of context, propagation of errors between agents, and unforeseen emergent behaviors.

To meet these challenges, cybersecurity must also become autonomous, capable of continuously detecting, isolating, deciding, correcting, and adapting.

The real challenge is no longer to keep humans at the center of every loop, but to design autonomous systems whose loops remain controllable, explainable, and reliable.

It is this dizzying prospect of autonomy that the INCYBER Europe 2027 Forum invites the whole world to explore.