Why most enterprise AI projects fail—and how to fix them

Thales Teixeira, a Professor of Practice at UCSD, has highlighted common pitfalls in enterprise AI adoption. He warned that poorly implemented systems can disrupt operations rather than improve them. His comments came during a discussion on how organisations often mishandle AI initiatives. Teixeira used the example of McDonald’s drive-through chatbot to show how immature AI can cause problems. He explained that systems not suited to their environment can create new bottlenecks instead of solving them.

He argued that successful AI adoption starts with choosing the right business challenges. Many companies, he noted, focus on minor issues rather than meaningful ones. Teixeira stressed the need to assess both the importance of the problem and the maturity of the technology before implementation.

Avoiding predictable errors in AI rollout was another key point. He emphasised that organisations frequently undermine their own AI projects by overlooking these factors. Teixeira’s insights point to a clear need for better planning in AI adoption. Selecting the right problems and ensuring technology readiness can prevent operational disruptions. His advice aims to guide businesses toward more effective and sustainable AI strategies.