Moving Beyond Deep Work to Generative Work
In 2016, Cal Newport gave us a powerful framework for navigating the modern world of knowledge work. He named two modes of working:
Shallow Work: “Noncognitively demanding, logistical-style tasks, often performed while distracted. These efforts tend not to create much new value in the world and are easy to replicate.”
Deep Work: “Professional activities performed in a state of distraction-free concentration that push your cognitive capabilities to their limit. These efforts create new value, improve your skill, and are hard to replicate.”
This distinction helped countless professionals rethink how they spent their time and inspired many to reclaim long, focused stretches of meaningful work. Newport’s Deep Work hypothesis was clear:
“The ability to perform deep work is becoming increasingly rare at exactly the same time it is becoming increasingly valuable.”
It was a call to put down the smartphone, silence the inbox, protect our minds from the creeping shallowness of digital distraction, and reclaim time, attention, and creative depth. And for many, me included, it worked.
But since then, something profound has been shifting in the nature of work itself.
With AI, The Game Has Changed
Today, the threat isn’t just that email and social media will keep us distracted with work that is shallow instead of deep. We now face the constant temptation to let AI do our thinking for us. Generative tools like GPT, Claude, Gemini, and others can now summarize, draft, rephrase, brainstorm, and even simulate thought processes. With just a few prompts, we can produce work that would have once taken hours of research, drafting, and revision.
This feels extraordinary and deeply unsettling in the same breath. Because while these tools save time, they can also short-circuit our thinking. A recent study at Stanford found that teams using generative AI produced more ideas, but fewer great ones. With output easier to generate, people were more likely to settle for whatever the model served up first.
Generative AI tempts us to hit the “easy button.” Why wrestle with a problem when the model can give you ten bullet points in seconds? Why struggle through a blank page when an AI can generate five opening paragraphs before your coffee is ready?
And so, ironically, while the pace of work accelerates, shallow thinking and an erosion in our ability to think deeply are even greater risks than ever.
Deep Work Still Matters, But It’s No Longer Enough
For that reason, Deep Work is still vital. It teaches us how to concentrate. It reminds us of the value of difficult cognitive effort, with the skills and intellectual capacities we develop when we engage in it regularly. And it offers a kind of creative satisfaction that no tool can replicate.
But Newport’s model was also built on two key assumptions: that if you want to create value in a knowledge economy, you need to “wring every last drop” of insight from your own mind, because there was no other mind available. And further, that the best way to do this was to minimize our exposure to a digital ecosystem that would distract us from doing so.
That’s no longer true.
We now work alongside generative minds. They’re not conscious or creative in the human sense, but they are fast, fluent, and increasingly capable. So, the question becomes: What kind of work should humans do when intelligent tools are everywhere? We need a new mode of work that builds on Newport’s insights, but one that adapts to a world where the most valuable worker is not just focused and thoughtful, but also tool-literate, discerning, and collaborative with AI.
Introducing Generative Work
Generative Work is professional activity in which humans lead and collaborate with AI to offload routine cognition, conserve mental energy, and direct their focus toward reflection, synthesis, and uniquely human insight that creates value which is more than the sum of its human and AI parts.
It’s not shallow work, where we coast on autopilot. It’s not traditional deep work, where we go it alone.
It’s a third path that blends human judgment with machine speed. It allows us to think deeply about the right things, while letting go of the tasks where machines can now outperform us in speed or volume.
Instead of spending our time pulling together the arguments of others, we use AI to assemble the raw material so that we can spend more of our brainpower crafting new arguments of our own. Instead of burning hours drafting boilerplate, we let the model get us to a rough draft, so we can devote ourselves to nuance, creativity, and ethical reflection.
Imagine a researcher using a language model to summarize a dozen papers, then spending their time identifying gaps in the literature. Or a policy analyst who prompts an AI to generate opposing arguments, then refines their final position with care. These are not shortcuts away from thinking. Instead, they are scaffolds that make space for thinking.
Why This Matters Now
As AI continues to transform how we work, write, think, and learn, Generative Work offers a productive response to this shift. It says:
Yes, conserve your mental energy.
Yes, work faster and smarter.
But also: think for yourself.
Don’t outsource your insight.
And above all: don’t confuse generation with understanding.
Generative Work may enable us to produce more. It must also enable us to produce better, and that means using the tools strategically, not passively.
It means knowing:
When to delegate to AI, and when not to.
How to think and work deeply first, prompting AI second
When to delay AI involvement to avoid short-circuiting your own insights
How and when to challenge AI, and when to have it challenge you and your teams
How to lead and collaborate with AI like a junior collaborator, not a final authority.
Deep Work was designed for a world where humans carried all the cognitive weight. But we now share that load. Generative work is about discerning where deep thinking is needed for the task at hand, and where it isn’t. With the right approach, we can avoid the pitfalls of shallow work without isolating ourselves from AI tools. And, we can take the appropriate steps to ensure that what we create is greater than the sum of its human and AI parts.
Check back for more on actionable ideas about how we can make the shift to Generative Work.
