Key takeaways ▼
Gleb Tsipursky, PhD, is the CEO and co-founder of Disaster Avoidance Experts, a consulting and training firm focused on helping organizations adopt AI through employee engagement and practical workflow redesign. He is a behavioral scientist and the author of The Psychology of AI Adoption at Work: From Resistance to Results (Georgetown University Press, 2026).
Tell us about your company?
I co-founded Disaster Avoidance Experts with our COO, Agnes Vishnevkin. We help organizations use the AI tools they already have to solve expensive, frustrating problems in their daily workflows. The work combines leadership strategy, hands-on employee training, and behavioral science. Teams learn to build and test tailored AI assistants without depending on a software development project for every improvement. Our consulting experience spans midsize organizations and major enterprises, including Applied Materials. We approach semiconductor applications with the same emphasis on verifiable results and clear responsibility for decisions.
What problems are you solving?
Many leaders have bought AI licenses but cannot point to corresponding changes in productivity, quality, or decision-making. Employees may be uncertain about where AI helps, wary of errors, or concerned about what using it means for their jobs. Others quietly use unapproved tools. We help managers understand these different responses, identify tasks worth changing, and give employees time and support to redesign the work. For higher-consequence decisions, we build in evidence checks, human review, and escalation rules before scaling adoption.
What application areas are your strongest?
Our strongest work is organizational AI adoption: identifying high-value use cases, developing staff capability, and embedding useful tools into actual workflows. Examples include researching and synthesizing complex records, preparing first drafts, creating internal knowledge assistants, and automating repetitive administrative processes. In semiconductor companies, I see particular potential in engineering knowledge retrieval, cross-functional handoffs, training, and analysis support. Decisions affecting process recipes, equipment control, or product disposition demand a much stricter validation and authorization process. I would begin with constrained tasks and expand authority only as evidence justifies it.
What keeps your customers up at night?
They worry about spending heavily on AI without seeing a return. They also worry about losing control of sensitive information, trusting an inaccurate answer, and pushing employees toward shortcuts that supervisors cannot see. Technical leaders face another concern: a system can produce plausible recommendations that fail when operating conditions change. Across industries, the central question is how to improve speed while preserving accountability, domain expertise, and workforce confidence.
What does the competitive landscape look like and how do you differentiate?
Organizations can work with software vendors, system integrators, and consultants who build custom AI applications. We focus on the adoption work that often receives less attention after a purchase: selecting real workflows, involving employees in design, teaching them to maintain useful tools, and defining how to verify outputs. We work within the platforms clients have approved rather than requiring a particular new AI vendor. Our behavioral-science background helps us diagnose why a technically capable solution might still fail to become standard practice.
What new features/technology are you working on?
We are refining reusable AI assistants, no-code automation workflows, and training methods that let nontechnical employees adapt tools to their own jobs. An equally important part of that work is evaluation. We want people to know how to test an assistant against realistic examples, recognize when it lacks reliable evidence, and route consequential decisions to the right reviewer. As agentic tools improve, those habits will determine whether organizations can delegate more work safely and productively.
How do customers normally engage with your company?
We usually begin with a discussion with senior leadership about goals, constraints, and where current AI efforts have stalled. We then gather input from employees and managers, select concrete use cases, and conduct practical workshops in which participants build and test tools for their own work. Depending on the organization, we also help design adoption roadmaps, manager support, and follow-up measures for usage, output quality, and time saved. A useful engagement leaves the client with capable employees, working examples, and a clear next set of decisions.
Contact Disaster Avoidance Experts
Also Read:
Share this post via: