My daughter Diana is sharp, and as a marketing and events professional at Vaco, she pays attention to what I publish. Recently, she asked me a question I have not been able to put down.
Other WRAL Top Stories
“Dad, everything you write about AI is the upside: fluency, adoption, getting on board. Why don’t you spend more time on the backlash — the things people are afraid of?”
I told her there’s no money in fighting the AI tsunami. My job — the work I do with executives and boards — is to help people thrive in an AI economy. That means spending my energy getting people technologically fluent and ready to make crucial AI adoption decisions.
I stand by that answer, but it was incomplete.
Because she was really asking this: “How can you help someone navigate a landscape if you only describe part of it?” You cannot coach a sailor by describing the wind and never the rocks. The whole point of fluency is judgment — knowing what the technology does well and knowing exactly where the hazards are.
So let me give a better answer. Let’s talk about the rocks and how to navigate around them.
AI anxiety is real, even though fears are unfounded
Job displacement is the rock everyone fears most. Challenger, Gray & Christmas attributed nearly 55,000 job cuts to AI in 2025. Goldman Sachs projects that entry-level knowledge workers are most exposed to new AI deployments — that is my daughter’s generation.
Here’s the nuance that matters to leaders: 63% of U.S. workers believe AI will decrease overall job availability. Yet every major projection, from Goldman to the World Economic Forum, counters that opinion. Thanks to millions of new roles created because of AI, analysts expect a net gain of 78 million jobs by 2030.
The gap between what people feel and what the data shows is an opening where fear and anxiety creep in. It’s up to leaders to bridge that gap with targeted training and clear communication. When employees gain valuable new skills, they worry less about job security. The same happens when leaders emphasize the importance of ongoing human oversight of AI outputs.
Without support, AI adoption can increase job stress
According to the TELUS Health Mental Health Index, which surveys roughly 19,000 workers globally, only 17% of workers believe AI will reduce their stress — and that small, optimistic group scored markedly higher on mental health than the worried majority. Read the inverse: for most of your workforce, AI is a source of stress, not a productivity tool.
These findings are clarified by Oliver Brecht, Vice President of the Workplace Options Center for Organizational Effectiveness (COE), who guides leaders on how to measure and manage the workplace conditions that affect mental health.
“As organizations adopt AI, they may unintentionally introduce psychosocial hazards: concerns about organizational justice, job insecurity, cognitive job demands, and the challenge of adapting to new systems,” he says. “Left unaddressed, this risk can lead to psychological injuries, erode trust, and undermine the very benefits AI promises.”
A study published in Frontiers of Public Health found that in AI-forward workplaces, employees face higher stress, work intensification, job insecurity, reduced autonomy, and blurred work-life boundaries.
Importantly, the researchers also identified significant benefits, such as increased efficiency, flexibility, and professional development. The key to lower stress and greater confidence was adequate training and organizational support.
A decision process without psychological safety
Here is the risk that should worry leaders most: relying on AI-generated content that hasn’t been checked by humans. AI models are prone to generating false citations, fictional figures, and incorrect analysis that closely resemble facts. And when AI scales faster than an organization's capacity to use it with judgment, mistakes happen.
In the Harvard Business Review, Jayshree Seth and Amy C. Edmondson describe how sustained AI use can undermine professionals' confidence in their ability to challenge AI results, even when they are experts. This is why psychological safety is an operational requirement. People only challenge results and reveal concerns when they feel safe to do so. When employees trust their leaders and the organization, they will identify issues before they turn into expensive mistakes.
It’s up to leaders to navigate the rocks
The stakes are high. In 2025, global enterprises poured roughly $684 billion into AI. By year-end, more than $547 billion — over 80% — had delivered no measurable business value, according to RAND’s analysis. These initiatives rarely fail for technical reasons. Instead, many are the result of human factors: lack of adequate training, unclear objectives, and employee resistance.
The journey ahead requires leaders to name the obstacles directly and create an intentional path around them. Talk about the AI anxiety and stress that your workforce is experiencing. Acknowledge their feelings in town halls, one-on-ones, and team meetings. Provide adequate training to increase AI proficiency and reduce anxiety. Communicate frequently about your Employee Assistance Program and other support services.
Build psychological safety into your AI adoption from the beginning. Ask your team where they don’t trust the tool. Create explicit procedures for people to voice concerns, challenge outputs, and admit confusion. Reward the person who says “I don't trust this result” before an inaccurate AI result becomes a problem.
Putting these safeguards in place can relieve the emotional pressure that a monumental change like AI transformation can place on employees.
Back to Diana
I think about Diana's question differently now. She wasn’t asking me to stop talking about the benefits of AI. She was asking me to tell the whole story — the wind and the rocks, the opportunities and the risks.
I compare AI to teaching Diana how to drive. A car increased her freedom, but it also exposed her to dangers. We didn't head straight to the highway. Instead, she gained experience in parking lots and on surface roads. You don't keep someone off the road forever, but you don't hand over the keys on day one. You build the skills that let them handle the risk — and then you let them drive.
Job insecurity, higher stress and AI hallucinations are real obstacles to an organization’s successful AI strategy. But it’s a mistake to let these common concerns prevent a company from investing in AI tools that can add efficiency and profitability. With any AI strategy, awareness is the key. Knowing where the risks are enables organizations to map a course around them. That’s how everyone thrives in an AI economy.
About the Author
Donald Thompson is Managing Director of the Workplace Options Center for Organizational Effectiveness, host of the High Octane Leadership podcast, and author of The Employee Engagement Handbook.