Why do AI projects fail? Spoiler: It’s rarely about the technology. The real barriers are organizational, cultural, and human. In this talk, we’ll uncover the reasons AI initiatives stumble and explore how to overcome them—through a mix of anecdotes, real-world failures, and actionable best practices.

We begin with a big lamenti—an unfiltered dive into the messy, chaotic world of AI project implementation. From tangled data silos to misaligned expectations, we’ll uncover hard truths through real-world examples that many will recognize: • The missing data problem: Sometimes it’s the smallest datasets that hold the biggest answers. • Passwords might be in retirement. • Decisions in the dark: When untrained people must make calls on things they don’t fully understand. • The shiny facade: Expectations, the pressure for quick wins, and the relentless demand for good news.

To go beyond personal experience, I’ll also bring fresh insights from a survey of my expert network closer to the conference. This will offer a broader, more representative perspective on why AI projects stumble—and how to fix them.

Timed Outline (45 minutes):

1. Introduction (5 min)
   • What qualifies me for this talk.
2. The Big Lamenti – Real-World Failures and Anecdotes (20 min)
   • Anecdotes that illustrate the challenges in a candid yet engaging way.
   • Learnings from the survey
3. Solutions and Best Practices (15 min)
   • Mindset: How AI is perceived and approached within organizations.
   • Data: The role of accessibility, governance, and quality as foundational pillars.
   • Collaboration: Bridging silos between teams, leadership, and departments.
   • Scale: The progression from small, focused initiatives to sustainable implementation.
   • Expectations: Aligning vision, goals, and outcomes for realistic project success.
4. Takeaways & Closing Thoughts (5 min)
   • Summary of insights and practical steps to avoid common pitfalls

• Key mindset shifts for sustainable AI success

1. Q&A (5 min)