Models Already Win — What Comes Next

Sean Goedecke writes that AI models already surpass humans on most evaluation tasks, and what's left is to win at something different. It's not a hype piece — it's a map of what comes next.

The premise is simple: if the model already does the base task, competing on that task is losing. You need to find another terrain — where the human has a structural advantage, not just because they're human.

This hits us directly. In the community, la gente is starting to use AI for everything — invoices, emails, translations. But the real advantage isn't in copying what the model does, it's in what the model can't replicate: relationships, context, trust.

Why this matters for us: el futuro no es competir contra la máquina en las mismas tareas — es encontrar lo que solo la gente puede hacer y cobrar por eso.