Some of the most interesting work in this space is in the “shared” memory models (in most cases today, vector db’s). Agents can theoretically “learn” and share memories with the entire fleet, and develop a collective understanding & memory accessible by the swarm. This can enable rapid, “guided” evolution of agents and emergent behaviors (such as cooperation).

We’re going to see some really, really interesting things unfold - the implications of which many haven’t fully grasped.