Which parts of your role are moving from human-led to AI-led?

One AI statistic from Anthropic caught my attention.

As of August 2026, Anthropic says Claude leads 26% of its AI R&D work.

“Leads” here means AI can complete most of the task end-to-end from a high-level prompt, with a human supervising.

More than 90% of its R&D work is already at least at the “AI collaborates” level.

That changes the question I think we should be asking about AI and jobs.

Not: “Will AI replace this role?”

But: “Which parts of this role are moving from human-led to AI-led?”

Because when AI takes over more of the execution, the human contribution doesn’t necessarily disappear.

My bet is that the work increasingly shifts toward defining the problem, setting direction, reviewing outputs, handling ambiguity and deciding what actually matters.

The interesting part of the AI transition may not be fewer tasks. It may be different tasks becoming valuable.