Most conversations about AI focus on efficiency: How much faster can people work? How many tasks can be automated? How much productivity can AI unlock?

But there is a growing risk hidden inside these questions. When technology accelerates work, it also accelerates decision-making, information flow, client expectations, and the cognitive demands placed on employees. I have noticed this in my work with teams across industries, many of whom report feeling worn down by the uncertainty and pace of change. I’ve also noticed that focusing on efficiency alone is not enough. The leaders who succeed with AI integration will be those who ask a different set of questions—ones that focus not only on technology, but also on the conditions people need to thrive alongside it.

AI and Sustainable Performance—What the Research Reveals

AI doesn’t just change how work gets done—it changes what humans must do to perform well. While AI may create important efficiencies, leaders also need to be aware of the ways in which AI use may challenge cognitive capacity, care, and connection.

A newly published study introduced the concept of AI brain-fry, which is a state of cognitive exhaustion from managing too many AI tools at once. Researchers discovered that cognitive overload was more likely in those professionals who regularly manage more than three AI tools at the same time.

Another study found that the most productive AI users are also 88 percent more likely to be burned out, disengaged, and twice as likely to quit. That same study found that 90 percent of workers see AI as a co-worker, 67 percent trust AI more than their colleagues, 64 percent say they have a better relationship with AI than with their human teammates, and 54 percent say AI is more empathetic.

Separately, researchers looked at how AI tools changed work habits over an eight-month period in a tech company. This company did not mandate AI use, and they discovered that employees worked at a faster pace, worked longer hours, and took on a broader scope of tasks. That seemed like a win until they also discovered scope creep, work slop (AI-generated output that fails to move a project forward and adds extra cognitive and emotional load onto colleagues who must fix or redo it), constant pressure to produce more, and loss of recovery time and deep thinking.

The researchers pointed out that over time, this overwork can impair judgment, increase the likelihood of errors, and make it harder for leaders to tell the difference between genuine productivity and unsustainable intensity. What’s needed are clear AI practices—clear norms and routines that add structure to how AI is used.

AI and Workload Sustainability: 5 Questions Leaders Should Ask

Work performance and sustainability concerns also need to be addressed in combination with technology integration and governance. As AI use increases, leaders and busy professionals should keep these five questions top-of-mind:

How do AI-fueled workflow changes impact workload sustainability, cognitive capacity, and burnout risk?
What are the psychological safety implications of hybrid human + AI teams?
How can leaders redesign roles to protect well-being and connection as automation increases?
What are the team communication norms that prevent overload in AI-heavy environments?
How will roles need to be re-designed or re-clarified?

Here is a summary of the potential work sustainability risks created by AI:

Cognitive overload: Constant new tools and updates
Increased workload: More review/correction tasks and expectations
Job insecurity: Fear of replacement by automation
Role ambiguity: Unclear performance standards
Work-life blurring: Always-on tools and alerts
Social isolation: Reduced human interaction
Skill pressure: Upskilling demands without support
Productivity expectations: Higher output expectations

In fact, a newly published meta-analysis reviewed 60 years of role stressor research (across 515 studies and almost 800,000 people) and discovered that these three stressors drive significant depletion at work:

Role ambiguity: Not knowing what one is supposed to do
Role conflict: Role expectations that contradict or do not align with actual duties
Role overload: Too much to do and too little time

While all three negatively impact both individual and organizational outcomes, researchers found that role ambiguity tended to be the most detrimental driver. What helps is for teams to have a clear purpose, clear decision rights, and a measure of autonomy over the size of their workload.

Ideas to Help Build Resilience Into Workplace Systems

Here are some additional ideas to help leaders and busy professionals build sustainability and resilience into workplace systems:

Build recovery into workflows. It’s common to go from one task or project to the next, but schedule recovery time before the next sprint.
Cross-train roles and responsibilities. I interviewed a partner at a large law firm who told me how important it was to the success of his practice for his clients to know his team. He explained that as a leader, internal and external clients may call him first or see him as the go-to person, but clients also need to feel comfortable with others on the team. It gives clients another attachment point and creates opportunities for team members to gain valuable experience interacting with clients (which also is a pathway to judgment development, something that junior knowledge workers will need to more intentionally develop in the coming months and years).
Recognize leaders and professionals who are preventative, not just reactive. Who spots problems early, manages risk quietly, and improves workflows so that breakdowns don’t happen?
Make sure professionals have decision-making discretion when appropriate. Employee empowerment is a resilience tool.
Protect small buffers. Add in a little extra time for a deadline; bake in some extra budget. You’re designing for recovery, so you don’t have to be as reactive if something doesn’t go as planned.

The AI conversation has moved from experimentation to execution. As leaders look for a competitive advantage, it is tempting to focus solely on speed, efficiency, and output. But sustainable performance should be treated as a business imperative, not an afterthought or something that people will “just figure out.” The organizations that gain the greatest return from AI will be those that invest as intentionally in human capability as they do in technology. The future of work will be shaped by leaders who know how to create the conditions for people and technology to perform at their best—together.