In mid-February, after coming back from China and catching up with AI, a partner told me about all the agents he had set up, the scaffolding, the workflows. He was a few days ahead of me and more applied. I listened and said: at that pace, in 2-4 months we will burn out.

Last week at #DataDrivenNYC, AI burnout was tangible in two ways. One, the presenters were not as confident as usual; realistically they implied we have no clue, but we are betting this will stick. Two, the crowd was not there either: the discussion was not about what they were doing, but about what the agents were doing. How quickly will we, the pioneers, become irate about AI due to an addictive, curiosity-driven dopamine burnout? How quickly will we understand what stays the same and frame the new world around that, so we have some sense of control?

I say this in light of the Coinbase cut this morning and the proposed structure with 15 direct reports. The cognitive load needed in an AI era for that to work is inhumane. Maybe a well-skilled agent can handle that, but a human? Don't bet on it.