Key Quotes
There's a certain bit of a split between what agents can do and what they're capable of, versus the amount of trust and confidence individuals and enterprises can place on those agents. Sometimes there is this bifurcation in the path where you say, let me go with agent capability, sacrificing trust, versus, let me go with trusted agents, even though I might have to sacrifice some capability. That's where we stand today, but that's not how it should be. You can have it all. You can have trusted agents that provide the maximal capability from the best frontier model out there, and you should not have to choose.
What's happening is that since the productivity unit is shrinking, the coordination surface and the coordination tax is exploding. Smaller teams means intra-team coordination is tending towards zero, but inter-team coordination has just exponentially exploded, because now we have an N squared problem.
Key Takeaways
Connecting agents is not the same as coordinating them. Open standards like MCP, A2A and AGNTCY let agents discover each other and pass messages, but they still can't reliably align on intent, agree on terminology or negotiate, so plan for coordination as a separate problem from connectivity.
Smaller teams make coordination between teams harder. A tiny team with tokens (one to five humans plus agents) works because judgment, taste and accountability stay with people, but with many small teams you need a shared context layer and named people who keep intent consistent across teams.
Judge agents on “better”, not just faster and cheaper. Token costs and speed are easy to measure, but you also need to check that agents are aligned, grounded and completing the task, so add outcome checks alongside your cost and latency dashboards.
