LLMs versus deterministic enterprise truth

Before enterprises can deploy AI agents that actually work, they need something most organizations don't have: a single, authoritative source of truth. Joe DosSantos, Workday’s VP of Enterprise Data and Analytics, joins Robb and Josh for a wide-ranging conversation about canonical knowledge, the semantic layer, and why data governance, a concept from the 1990s, has suddenly become essential for AI deployment.

Large language models are predictive engines modeled to anticipate what users probably likely mean. For B2C applications where multiple interpretations are acceptable, this works fine. But enterprises need deterministic truth, not probabilistic guesses. The trio outline a solution in three layers: establishing canonical knowledge, building a semantic layer to translate between human definitions and machine-readable formats like YAML, and using LLMs as an interface to deterministic back-end systems.

For leaders evaluating AI investments, this episode clarifies what actually needs to be built before agents can deliver value: not flashy use cases, but the unglamorous, essential work of data governance and semantic translation.

---------- Support our show by supporting our sponsors!

This episode is supported by OneReach.ai

Forged over a decade of R&D and proven in 10,000+ deployments, OneReach.ai’s GSX is the first complete AI agent runtime environment (circa 2019) — a hardened AI agent architecture for enterprise control and scale.

Backed by UC Berkeley, recognized by Gartner, and trusted across highly regulated industries, including healthcare, finance, government and telecommunications.

A complete system for accelerating AI adoption — design, train, test, deploy, monitor, and orchestrate neurosymbolic applications (agents).

- Use any AI models

- Build and deploy intelligent agents fast

- Create guardrails for organizational alignment

- Enterprise-grade security and governance

Request free prototype:

https://onereach.ai/prototype/utm_source=soundcloud&utm_medium=social&utm_campaign=podcast_s7e2&utm_content=1

---------- The revised and significantly updated second edition of our bestselling book about succeeding with AI agents, Age of Invisible Machines, is available everywhere:

Amazon — https://bit.ly/4hwX0a5

Chapters -

0:00 – Welcome to Invisible Machines

1:28 – Why AI Agents Fail Without a Source of Truth

2:34 – Canonical Knowledge Is More Than Feeding Data to an LLM

3:16 – LLMs Are Good at Language, Not Truth

4:16 – The Convergence of Governance and Generative AI

5:48 – Implicit vs Explicit Knowledge Explained

7:31 – Why Accuracy Breaks Down in AI

8:37 – The Real Launchpad for AI: Get the Facts Right

9:42 – Alignment, Not Intelligence, Is the Hard Problem

10:53 – Semantic Layers: Teaching Machines Meaning

12:38 – LLMs Are Interfaces, Not Systems

14:26 – Routing Questions: Inference vs Deterministic Answers

16:21 – Canonical Knowledge Requires Human Ownership

18:16 – There Is No ROI for Data (It’s the Foundation)

23:59 – From Use Cases to Systems Thinking

Episode Credits:

Robb Wilson - Host

Josh Tyson - Host

Elias Parker - Executive Producer

Vishal Menon - Producer

Maksym Zlydar - Audio/Video Editor

Mykhailo Lytvynov - Audio/Video Editor

Eugen Petruk - Graphic Design

Alla Slesarenko - Copy

Vira Prykhodko - Web Development

#InvisibleMachines

#Podcast

#TechPodcast

#AIPodcast

#AI

#AgenticAI

#AIAgents

#DigitalTransformation

#AIReadiness

#AIDeployment

#AISoftware

#AITransformation

#AIAdoption

#AIProjects

#EnterpriseAI

#CanonicalKnowledge

#DataGovernance

#SourceOfTruth

#AIArchitecture

#DeterministicAI