Gleb Tsipursky on the Psychology of AI Adoption

August 27, 2026
Gleb Tsipursky · Disaster Avoidance Experts
CEO · Disaster Avoidance Experts
AI InterConnect Podcast · Published August 27, 2026

Dr. Gleb Tsipursky is the CEO of Disaster Avoidance Experts, which helps organizations overcome resistance to AI and build internal capabilities so employees can create and improve AI solutions around their own workflows. A behavioral scientist and author of a forthcoming Georgetown University Press book on the psychology of AI adoption at work, he tells Laurent Cochet why the obstacle to AI is psychology rather than technology, how to measure results, and what leaders can do in the next thirty days when an AI initiative stalls.

From cognitive bias research to AI adoption

Tsipursky traces his interest to the dot-com era. The same executives praised by top newspapers in 1999 were criticized in 2002, though their decision-making style had not changed; only the circumstances had. That convinced him that decision making, adoption and behavior mattered more than most people realized.

He earned a PhD in behavioral science at UNC Chapel Hill, taught there and later at Ohio State, and consulted for companies on the side. His new book is peer reviewed and, he says, draws on over a hundred consulting projects on AI adoption, including over six hundred focus groups and thousands of survey responses.

His central argument is that leaders pattern-match AI to earlier rollouts, such as customer relationship management (CRM) or enterprise resource planning (ERP) software, where the challenge was learning new screens and habits. AI, he says, is surprisingly easy to learn because you can ask a tool like Claude how to use it for claims, underwriting or invoice reconciliation. It is not the technology, he says; it is the psychology.

Building capability instead of buying wrappers

Disaster Avoidance Experts urges companies to avoid dependence on expensive, inflexible AI vendors. Many vendor products, Tsipursky argues, are AI wrappers that harness a model behind the scenes, something employees can recreate with broad platforms like Claude, ChatGPT, Copilot or Gemini, customized to their own workflows.

That stance creates a business tension. Most of the firm's work happens at the start: assessing which blockers are present, then training and advising leaders. After that, clients need little help beyond complex builds or expanding a pilot to another department. By his account, building AI tools for clients is around 3% of revenue, and the rest comes from helping them adopt and use AI effectively.

The hardest part, he says, is constantly finding new clients; referrals from satisfied clients help. Delivering the transformation is simple by comparison; educating the market is difficult because large vendors have much stronger media and marketing teams.

Measuring efficiency and effectiveness

Tsipursky cites an MIT study showing that 95% of AI pilots do not return the resources invested, and a study published in Harvard Business Review on AI work slop, the problematic output people spend time fixing, which he puts at $186 per employee per month. Leaders, he says, need that wider context and the right metrics.

He separates two kinds of success metric. Efficiency is doing the same workflow with less time, ideally measured on identical work; a tax preparation firm he trained compared how long the same client's taxes took last year and this year. Effectiveness is getting better outcomes from the same resources.

His effectiveness example is a wealth management firm he trained to build a tool to scan its local area for public events, such as a business sold for $15 million, that leave someone with money to manage. The tool finds contact details and sends automated outreach, turning cold messages into inbound conversations so the firm only deals with interested prospects.

Three blockers and the next thirty days

For a CEO whose AI initiative works technically but has stalled, Tsipursky's advice for the next thirty days starts with an anonymous survey of concerns, anxieties and priorities. In chapter six of his book he describes three groups. AI alarmists fear that AI will take their jobs. Pragmatic resistors are less worried about jobs but feel AI threatens their professional identity.

The third group, reluctant adopters, use AI but hide it out of shame. He recalls a training for a peer group of executives where one CEO told a story about his VP of sales. A high-stakes annual presentation that usually took the VP about three days had been pulled together by the VP's executive assistant in one and a half hours using AI tools. The VP told the CEO privately and asked him not to tell the other executives, who would think it was cheating. When people hide gains like this, he says, individual productivity does not really translate into company goals.

Leaders should also ask employees which tasks they most hate, such as gathering information across systems, and train people to automate those. That sets a baseline for acceptable AI use and helps both reluctant adopters and pragmatic resistors; the alarmists take more work.

Growing through the book

Tsipursky sees the book as a platform for podcasts and conversations, though it has also crowded his schedule. Most clients now come through referrals, and his goal is to increase the client base by 20% from people who read or hear about the book. His wife, who previously worked in the back end of the business, is now extending the firm's work to nonprofits. He has also built fifteen AI agents to support his own work.

If the book succeeds, he says the firm will probably hire someone trained to build agents and serve clients. His main market is leaders of small and medium enterprises in professional services such as tax preparation, finance, insurance and legal services; he also works with manufacturing and retail.

Watch the full conversation

Watch the full conversation above. Tsipursky says a free sample of the book is available on the Disaster Avoidance Experts website, on its AI book page.

Watch on YouTube →

AI InterConnect hosts conversations with founders, investors and operators building with AI. Explore more episodes on the AI InterConnect podcast page.

Back to Blog

AI InterConnect

AI InterConnect connects founders, CEOs, investors and technology leaders through conversations, market intelligence and growth infrastructure.

Explore our podcast, research and GTM systems to identify opportunities, build authority and accelerate business growth.