The Real Reason AI Projects Fail: You Can't Sprinkle Fairy Dust on Old Processes

Jan de Vries · 2026-09-16

AI is moving into boardrooms faster than many businesses can figure out what to do with it. For Jim Harris, a specialist in GenAI and disruptive innovation, the bigger risk isn't moving too slowly—it's bolting new technology onto old ways of working and expecting transformational results. Harris advises executive teams and has worked with organizations like American Express, Barclays, IBM, SAP, Walmart, and the UK Cabinet Office. In 2024, he was named Speaker of the Year by TEC Canada for his work on AI. In an exclusive interview with the AI Speakers Agency, he explained why companies need to redesign processes from scratch, how leaders can reduce employees' fear of AI, and why keeping humans involved remains critical. ### The Fear Factor: How to Introduce AI Without Scaring Your Team When rolling out AI, leaders often overlook the emotional side. "First off, it's important to create psychological safety," Harris says. "The majority of people in developed nations actually fear AI. They fear job loss. As adults, we tend to fear things or are uncomfortable with new ways of working." So what should leaders do? Harris suggests a three-pronged approach: - Assure employees that their jobs are safe. - Provide training to help them adapt. - Emphasize that AI will enhance their skill set and employability. "Those are a few things organizations have to do," he adds. ### Why Sprinkling AI Fairy Dust Doesn't Work Harris shares a striking example from Ernst & Young (EY). With 400,000 employees and a billion dollars spent on AI, they had a risk management process that took 46 hours. They applied AI with the original process designers and got it down to 32 hours—a 20% gain. But then they tried something different: they paired tech-savvy new hires with risk managers who had never been involved in the process. The result? 15 minutes. After adding back compliance steps, it settled at six hours—an 80% reduction. > "You can't just sprinkle some AI fairy dust on your existing processes and hope to get the greatest gain." Harris's key takeaway: start with a blank sheet of paper. Ask, "If we were designing this process today from first principles, what would we do?" Don't try to enhance a 2018 process with AI—redesign it entirely. Also, bring in fresh eyes. People who designed the old process are often too attached to it. ### AI Beyond ChatGPT: Transformational Gains Harris points to Demis Hassabis, founder of Google DeepMind and Nobel Prize winner, as an example of AI's potential. While he didn't elaborate, the message is clear: AI can go far beyond tools like ChatGPT to deliver staggering improvements. The lesson? Don't just apply AI to old processes. Rethink everything. And remember, humans are still essential—especially for compliance and oversight.