In a recent episode of the Apple @ Work podcast, behavioral scientist Dr. Gleb Tsipursky joined to unpack what really holds back successful AI adoption in enterprises—not bugs, servers, or models, but people. He explored how psychological dynamics like fear, stigma, professional identity, and uncertainty often derail AI rollouts that were technically sound.
Key Psychological Barriers to AI Adoption
Tspipursky identifies three main categories of resistance employees bring to AI integration. First are those who fear AI will replace them—this fear of job loss triggers anxiety and pushback. Second are professionals whose sense of identity is tied to tasks AI may take over; they feel their value and expertise are under threat. Third are the silent adopters—those who use AI tools behind the scenes but hide it, fearing negative judgment or stigma from colleagues and leadership.
These barriers don’t collapse under traditional change-model tactics like training sessions or technology upgrades. Instead, they require nuanced interventions that address identity, belonging, and psychological safety.
Strategies That Make AI Work for People
Tsipursky argues that organizations need to reframe AI rollouts as human-first initiatives. Instead of launching AI by targeting high-value work, it can be more effective to start with tasks employees dislike—automating tedious work builds trust and opens space for broader adoption.
Leadership plays a critical role. Visible encouragement, early adopters serving as champions, transparency about decisions and expectations, and involving staff in choosing how AI is used all help shift organizational culture. Psychological safety becomes vital: people need permission to experiment, make mistakes, and provide feedback without fearing professional consequences.
Another element: sharpening judgment. As AI takes over more routine work—drafting, analyzing, summarizing—human roles grow richer in oversight and decision-making. Organizations that succeed will emphasize when people should trust AI output and when human scrutiny must prevail.
Measurement also matters. Rather than tracking metrics like number of users or usage hours, value should be tied to outcomes: error reduction, time saved, quality improvements, and how well the AI-enhanced work aligns with business goals.
Tspipursky’s new book, “The Psychology of AI Adoption at Work: From Resistance to Results” (Georgetown University Press, 2026), serves as a central framework for these ideas. He pulls together psychological research, case studies, and consulting experience to argue that technical tools are necessary—but insufficient—without cultural work.
Empowering middle managers also emerges as essential. These leaders often bear the risk of implementing executive direction yet carry less visibility. When they’re engaged, resourced, and serve as translators between executives and frontline teams, they can make AI adoption stick.
Finally, Tsipursky reminds us that AI adoption is not a one-time project—it’s an ongoing journey. As AI capabilities evolve and workflows shift, organizations will need to keep learning, iterating, and adapting what people, roles, tools, and standards are required.
Why this matters: as enterprises race to adopt AI, many will hit a wall not because of technology limits but because of real human concerns—job identity, trust, autonomy. Leaders who recognize this and act accordingly won’t just roll out tools—they’ll unlock real value, stronger teams, and sustainable transformation. Watch for firms that put psychological safety, employee agency, and continual measurement at the core of their AI strategy—they’ll define what success looks like going forward.
