Why AI Doesn’t Reduce Work — It Intensifies It (HBR Research Summary)

A solid piece of research worth your attention.

Below is a summary and translation of the key ideas from AI Doesn’t Reduce Work — It Intensifies It, published in Harvard Business Review on February 9, 2026, by Aruna Ranganathan and Xingqi Maggie Ye. The full HBR text is behind a paywall, but this summary accurately captures the core findings and conclusions.

The Promise vs. the Reality of AI Work Intensity

The expectation was clear: generative AI would handle routine tasks — drafting standard documents, summarizing information, debugging code — and free up time for higher-value work.

In practice, AI does not reduce workload. It intensifies it. Workers end up doing more tasks, faster, and across a broader scope than before.

Why AI Makes Work More Intense, Not Less

Task expansion. AI enables people to take on new types of work and fill knowledge gaps they previously couldn’t cross. Product managers start writing code. Researchers take on engineering tasks. The scope of individual responsibility expands — but time does not free up.

Blurring the boundary between work and rest. Because AI tools are always accessible, workers begin completing tasks during what used to be downtime — over lunch, between meetings, at the end of the day. Work becomes nearly continuous.

Increased multitasking and cognitive load. Working with AI means managing multiple streams simultaneously: crafting prompts, evaluating outputs, switching between tasks. This leads to mental fatigue, reduced decision quality, and heightened demands on sustained attention.

Consequences for People and Organizations

The effects compound across both individuals and companies:

• Rising work intensity: even when individual tasks are completed faster, the overall volume and pace of work increases.
• Cognitive exhaustion and burnout risk: employees report fatigue, difficulty concentrating, and a persistent sense of always being “on.”
• Erosion of work-life boundaries: work increasingly bleeds into personal time as the friction of starting a task drops to near zero.

What the Research Recommends

To avoid these negative effects, organizations need structural responses — not just individual coping strategies:

Establish an AI practice — explicit norms that govern how, when, and for what purposes employees use AI tools.

Build in structured breaks and sequential workflows to prevent an always-on mode from becoming the default.

Keep humans in the loop for reviewing AI outputs rather than treating the tool as a fully autonomous actor.

The Bottom Line on AI Burnout

AI can genuinely accelerate individual tasks. But without organizational guardrails and a rethinking of workflows, it:

• expands the volume of tasks and responsibilities people take on
• dissolves the boundaries of the working day
• amplifies multitasking
• creates the conditions for burnout

The net result: work doesn’t shrink — it goes deeper and gets more intense.