Beyond Burnout: Why Your Workforce Is Quietly Cracking

“It’s not that I’m lazy, it’s that I just don’t care.” 
— Peter Gibbons, Office Space

For fifty years, we built organizations for throughput: layers of bureaucracy, brittle processes, internal competition, and profit at all costs. Now we’re watching people break under systems designed to treat them like parts. That fracture has a name: “quiet cracking,” or a persistent unhappiness that erodes performance and fuels attrition, a term introduced by TalentLMS in 2025.

Quiet cracking isn’t the loud withdrawal of “quiet quitting.” It’s the inward snap that happens when people keep producing while losing connection to meaning, to teammates, and to themselves. In both my TEDx talk and Upwork’s latest research, one antidote keeps showing up as the essential balm to disengagement: meaningful connection to the work itself, to the people doing it, and to tools that amplify human purpose.

Our Tools to Teammates research makes the paradox plain. AI is boosting output — 77% of executives report gains, and employees self-report about 40% higher productivity. Yet the workers getting the most done with AI are also the most at risk: 88% report burnout, they’re twice as likely to consider quitting, 67% trust AI more than coworkers, and 64% say they have a better relationship with AI than with teammates. When technology becomes the most reliable “teammate,” that isn’t a tooling problem; it’s a culture problem.

So, no, AI won’t repair broken systems, and neither will forcing people back into the office. (If the culture doesn’t value connection, presence just concentrates the crack.) Also, most companies aren’t even equipping people to use AI well: only 1 in 4 offer formal AI skills training, even as usage accelerates. This is why the issue is systemic, not seasonal, and why the fix is a leadership mindset shift, not a software rollout.

Contrast that with the independent economy, especially Gen Z. Upwork’s Gen Z research shows many are choosing diversified, skilled careers: 53% of Gen Z freelancers work full-time hours across a portfolio of projects, and they’re more likely to hold postgraduate degrees than their peers who do not freelance. They are also out in front on AI. Sixty-one percent of Gen Z freelancers are adopting generative AI compared to 41% of their Gen Z full-time peers; 39% of Gen Z freelancers already hold an AI certification.

Crucially, intrinsic motivators — mastery, autonomy, relatedness — are met at higher rates among portfolio careerists and Gen Z business owners than in the general workforce. These workers feel more control over how work is organized and report a stronger connection to others, which is exactly what traditional systems are lacking. This aligns with the argument of my book, Essential: People are capable and creative, but we’ve trained those qualities out of work with scarcity, surveillance, and conformity.

If leaders want durable performance in the age of AI, the play is not “more dashboards” or “mandatory office.” It’s redesigning work for humans who now collaborate with increasingly capable systems. Here’s a concise blueprint:

• Redesign roles for Human + AI (not Human vs. AI). Shift from tool deployment to work redesign that builds autonomy, trust, and psychological safety so people can do higher-judgment work with AI as a teammate, not a replacement. Our research calls for designing work for humans + AI, and the risk data (burnout, attrition among top AI performers) shows why redesign, not speed alone, matters.
  
• Rebuild connection as an operating metric, not a perk. Treat relationship quality like a KPI. With AI perceived as a “coworker” and many high performers trusting it more than colleagues, connections must be designed into teams and measured alongside output.
  
• Invest in capability, not compliance. Close the training gap. Only about a quarter of companies provide formal AI training, while most workers use AI to augment their work. Capability building beats policy edicts every time.
  
• Tap frontier talent pools (freelancers, managed services, agencies). Don’t fix yesterday’s org chart. Stand up mixed teams that add independent experts already fluent in human + AI workflows. Demand is surging — searches on Upwork for talent skilled in working with AI agents are up nearly 300% — and freelancers report strong skill growth and positive career impact from AI.
  

Quiet cracking is not a worker failure; it’s a system failure. AI won’t save a culture that treats people like components, and office mandates won’t manufacture meaning. The solution is leadership that chooses trust over surveillance, development over directives, and connection over control. If you’re serious about performance, redesign the system so humans can thrive with AI. The cracks are showing you exactly where to begin.