Europe 2026
Most Enterprise Agentic Projects Are Doomed, Here's Why — Jess Grogan-Avignon & Jack Wang, Accenture
Jess Grogan-Avignon , Jack Wang , Accenture
Overview
Most enterprise agentic AI projects are destined to fail due to fundamental tensions between traditional enterprise structures and the demands of machine-speed AI development. Enterprises, built for human pace with layers of control and process, struggle to adapt to the rapid iteration and emergent behaviors characteristic of AI. This talk outlines five key tensions—speed, value, delivery, trust, and moat—and offers a framework for navigating them to achieve successful AI adoption.
Who should watch
- AI Engineers and Builders
- Product Managers and Leaders
- Enterprise Architects
- Anyone involved in deploying AI at scale within large organizations
- Those facing challenges with slow AI project timelines and ROI
- Teams struggling to integrate AI with existing enterprise processes
Key takeaways
- Traditional enterprise scaffolding, designed for human speed and control, acts as a significant drag on AI adoption, hindering the ability to capture value at machine speed.
- Enterprise finance models, focused on upfront certainty and predictable ROI, are ill-suited for AI projects where scope and value are often discovered through experimentation. A venture capital-like portfolio approach is recommended.
- Agentic delivery requires a shift from traditional software development to a hypothesis-driven approach, focusing on building statistical confidence through rapid build-evaluate-iterate loops rather than fixed requirements.
- Building trust in AI systems is paramount and should be engineered deliberately through progressive autonomy, starting with shadow modes and gradually increasing control based on evidence and outcomes, not just feature completion.
- An enterprise's true competitive moat lies not in its existing data or systems, but in its "living memory"—the ability to rapidly compound learning from customer interactions and feedback loops.
- Every shipped feature should either generate feedback or act upon learned insights; otherwise, it risks being easily copied. Feedback is presented as the only sustainable moat.
- The core prescription involves adopting a VC-like investment mindset for AI bets, upgrading processes for machine speed, and engineering for trust with continuous feedback loops from the outset.
Notable quotes
*The very thing that has made these companies so successful, which is increasingly becoming the drag, the thing that is holding them back from capturing AI value at scale.*
*The real tech debt here goes beyond the legacy code that exists within applications. And it's the years of underinvestment in the engineering automation, CI/CD, et cetera that allows companies to move faster while maintaining control.*
*Your moat is not in what you hold from yesterday. It is what it is in what you are learning and compounding every day.*
Unofficial community note. Prefer the recording for nuance.