77% Of Employees Report AI Has Increased Workloads

A large share of employees say AI has made their jobs harder rather than faster: 77% report increased workloads and productivity challenges after AI became part of day-to-day work. The tension at the center of the findings is a mismatch between what leaders expect AI to deliver and what workers experience—especially when “efficiency gains” quietly become higher output demands.

The discussion frames the problem as a new baseline for performance. If AI helps people complete tasks more quickly, managers may treat that speed as the new norm for everyone’s output. That can turn a 40-hour job into something that feels like 80 hours in practice: workers may still be producing more, but they’re judged against an inflated expectation. In that scenario, taking breaks or working at a slower pace can be misread as underperformance, even when the worker is actually doing more than before.

The numbers behind the mismatch are stark. A separate set of results cited in the conversation reports that 96% of C-suite executives expect AI to boost productivity, while many employees say they don’t know how to achieve the gains their employers anticipate. Nearly half (47%) of employees using AI report they can’t reach expected productivity improvements, and 40% say their company is asking too much when it comes to AI-enabled work. The strain shows up in retention and burnout signals: one in three full-time employees says they’re likely to quit within six months due to feeling overworked and burned out. Burnout is also reported broadly, with 71% of full-time employees described as burnt out, alongside 65% struggling with employer productivity demands.

A key complication is that the productivity story differs depending on who is doing the work. The conversation points to claims that freelancers are meeting productivity demands—and sometimes exceeding them—while full-time employees feel the pressure more acutely. The reasoning offered is partly economic and partly motivational: freelancers are paid per hour, so extra output can directly translate into extra earnings, while full-time workers face fixed compensation and shifting expectations. That difference helps explain why organizations might respond to AI strain by leaning more heavily on alternative talent pools.

The cited research also argues that organizations are failing to unlock AI’s full value because they’re inserting AI into outdated work models rather than redesigning how work is organized. The proposed remedy emphasizes changes beyond the “tech stack”: bringing in outside experts, rethinking how productivity is measured, and adopting skills-based hiring and talent development. Freelancers are positioned as a lever for “AI-ready” capacity, with claims that many leaders use freelancers to jumpstart delayed AI projects and that freelancer use correlates with innovation outcomes.

Overall, the core takeaway is not that AI fails to improve capability, but that workplaces may convert capability into pressure—raising output expectations faster than workers can adapt—unless companies redesign incentives, training, and workflows to match the new reality of AI-augmented work.