Why 95% of AI Pilots Produce Zero ROI | Yasmeen Ahmad, Google Cloud

Unlocking Enterprise "Dark Data": BigQuery's Knowledge Engine and the Rise of Agentic AI

Yasmeen Ahmad, Managing Director of Product Management for Data & AI Cloud at Google Cloud, explains why 80–90% of enterprise data remains dark and how new tooling can surface the business context analysts have traditionally held in their heads. The episode focuses on BigQuery's Knowledge Engine and the semantic layer as the practical unlock for enterprise AI adoption in 2026.

She also explores how organizations scale autonomous AI agents from tens to thousands, why continuous evaluation is required to maintain trust, and what Google learned as 50% of its own code became AI-assisted. The discussion covers the common reasons 95% of AI pilots fail to deliver measurable ROI and how platform partnerships and governance can change that outcome.

Key Topics Covered:

• Why 80–90% of enterprise data is "dark" and the role of GenAI in unlocking it
• How BigQuery's Knowledge Engine captures invisible business context and semantics
• The semantic layer as the next major enterprise AI unlock, beyond model size
• Scaling autonomous AI agents from ~50 to 2,000 within organizations
• Intent-driven, agentic AI that executes outcomes rather than step-by-step tasks
• Replacing traditional unit testing with continuous evaluation for AI agents
• Google's internal AI adoption: 50% of code AI-assisted and measurable engineering gains
• Why 95% of AI pilots yield zero ROI and what changes that trajectory
• Framing governance as an accelerator—the "brakes" that enable confident, rapid execution
• The importance of culture and founder mentality over budget for AI success

Episode Timestamps:

00:00 - Introduction and welcome
00:50 - Being Scottish in Silicon Valley and the power of community
03:13 - The career thread: curiosity, pivots, and getting outside your comfort zone
06:20 - What makes data fascinating: the hidden stories inside numbers
07:48 - Why data is the lifeblood of enterprise AI
10:19 - 80–90% of enterprise data is "dark" and untouched
12:59 - What BigQuery actually does (explained simply)
14:50 - The invisible work: knowledge layers and business semantics
17:38 - The agentic AI moment: agents that think, plan, and execute
20:41 - From 50 to 2,000 autonomous agents inside enterprises
22:07 - Why you can't evaluate AI agents like traditional software
25:46 - Signals of AI readiness: Google's 50% AI-written code and Honeywell's 30% efficiency gains
30:21 - Why 95% of AI pilots produce zero ROI
35:37 - Governance as a speed accelerator, not a brake
39:53 - Who's best poised to win: culture over budget
45:59 - Why do you do what you do?

About the Guest:

Yasmeen Ahmad is Managing Director of Product Management for Data & AI Cloud at Google Cloud, leading product strategy around data, semantics, and agentic AI adoption. Her work centers on making enterprise data usable by capturing business context and enabling scalable, trustworthy AI systems.

About the Company:

Google Cloud provides data infrastructure and AI platform services, including BigQuery and the Knowledge Engine, designed to surface semantic context and unlock previously untapped enterprise data. The platform emphasizes scalable agent deployment, continuous evaluation pipelines, and governance frameworks that drive measurable ROI for enterprise AI initiatives.