The Rise of Context-Switching: How Agentic AI Upends the Era of Deep Work

As AI agents work in parallel, the new premium skill is not focus, but the ability to oversee and switch between multiple contexts.

Published on Feb. 21, 2026

Got story updates? Submit your updates here. ›

The article argues that the most valuable professional skill is no longer the ability to focus intensely for long periods, but rather the capacity for context-switching and overseeing multiple AI agents working in parallel. The author explains how AI has inverted the traditional emphasis on "deep work", as the new bottleneck is not attention but judgment across different contexts. The piece draws historical parallels to figures like Napoleon and Julius Caesar who could manage multiple tasks simultaneously, and suggests that teenagers adept at navigating various digital platforms are glimpsing the future of work.

Why it matters

This shift represents a fundamental change in how knowledge work is structured and valued. Whereas previous generations prized the ability to enter a state of "flow" and concentrate intensely, the new premium is on the capacity to fluidly switch between different contexts and oversee the work of AI agents. This has implications for how we educate, train, and organize knowledge workers going forward.

The details

The article explains how AI agents work in parallel, spawning other agents, iterating, and checking their work, while the human overseer waits and occasionally provides nudges or additional context. This is a stark contrast to the traditional focus on minimizing distractions and achieving uninterrupted "deep work". The author argues that the new scarcest resource is not attention, but judgment across multiple contexts. They draw parallels to historical figures like Napoleon and Julius Caesar who could manage multiple tasks concurrently, as well as modern teenagers who seamlessly navigate various digital platforms.

  • The author has been programming for over 2 decades and says this is the biggest change to their basic coding workflow in that time.
  • The shift towards context-switching and agentic AI has happened over the course of just a few weeks.

The players

Gloria Mark

A professor at UC Irvine who studies digital distraction and found that it takes over 23 minutes to fully regain focus after a single interruption.

Mihaly Csikszentmihalyi

A psychologist who gave us the concept of "flow" - the state of complete absorption where time dissolves and performance peaks. Flow is explicitly monotropic, requiring highly focused attention.

Joe Dumars

An NBA player who entered a state of flow during Game 3 of the 1989 NBA Finals, scoring 17 straight points and winning MVP without realizing he was being guarded by a 5-time All-Defensive team member.

Ben Sliney

The FAA's National Operations Manager who, on his first day on the job on 9/11, made the unprecedented order to land about 5,000 aircraft immediately, demonstrating the ability to orchestrate a complex system with incomplete information.

Got photos? Submit your photos here. ›

What they’re saying

“We kept waiting for him to miss. You could feel the whole building waiting. But it was as if he had forgotten how. He was scary.”

— Mitch Kupchak, Lakers assistant general manager (Sports Illustrated)

What’s next

The article does not mention any clear future newsworthy moments related to the story.

The takeaway

This shift from the era of "deep work" to one focused on context-switching and overseeing agentic AI represents a fundamental change in how knowledge work is structured and valued. The new premium skill is not intense focus, but the capacity for fluid judgment across multiple contexts - a skill that may come more naturally to younger generations already adept at navigating various digital platforms.