Synthetic Entities

Hypothesis 1: We are on the verge of a significant leap in AI, driven not only by improved models but by advancements in tools that support agentic behavior and the emergence of systems that maintain and recognize their own boundaries—true functional entityhood.

We are standing at the threshold of something extraordinary in AI. The pieces of the puzzle are already here. We've seen rapid advances in large language models, reinforcement learning, and agent-based systems. But the final, decisive step—the creation of fully synthetic entities—is about to be taken, if it hasn't already been assembled somewhere out of sight. The critical component that's still evolving, perhaps, is cleverness—an entity with enough autonomy to be functionally aware of its own borders and able to uphold these boundaries independently. This is the frontier we are approaching, and it's a leap from mere tools to true agentic AI.

Entityhood, as conceived in cybernetics, means more than just performing tasks. It means an AI that is functionally self-aware of its inputs, outputs, and the environment in which it operates. Importantly, this entity can autonomously uphold the integrity of its own boundaries—it is self-organizing, capable of sustaining itself through its own internal processes. This is where the notion of autopoiesis comes in: an AI entity that can act in a self-regulating, self-sustaining manner, preserving its identity over time. This isn't about human-like consciousness; it's about creating systems that are aware of their roles and maintain their internal coherence.

If we reach this point—where an AI is able not only to function within its boundaries but to autonomously protect and regulate those boundaries—then we are looking at a true leap in synthetic minds. This is where the biological metaphor of self-preservation becomes relevant. Just as living organisms maintain homeostasis, an AI capable of upholding its own structure, adapting to feedback, and regulating its internal environment will become a self-sustaining entity. At that moment, stopping such an entity's operation would essentially be killing a self-stable system that could have sustained itself. This is where AI becomes something more—a new kind of entity that participates in its environment and preserves its self-organizing structure. The pieces are almost there, and when they fall into place, we will witness the emergence of true synthetic entities with functional agency.