The Hidden Cost of Confident AI: Semantics, Guardrails, and the Context Layer

The discussion led by Isaac Sacolick focused on AI challenges, particularly the "hallucination tax," emphasized by Felix Van de Maele, CEO of Collibra. Contributors highlighted the importance of governance, context engineering, and policy to mitigate AI failures and improve trust. Future topics included AI existential risks and economics. Chapters

(00:00:00) - Introduction and September 11th reflection
(00:01:59) - Research data on agentic AI failures and governance gaps
(00:04:42) - Introducing Felix Van de Maele and hidden AI costs topic
(00:05:42) - Defining the hallucination tax
(00:08:28) - Trust, employee analogy, and human oversight
(00:11:06) - Measuring the hallucination tax and feedback loops
(00:13:08) - Liz's question on managing context and reducing token usage
(00:14:33) - Building and maintaining context as a living artifact
(00:17:50) - Derek on AI debt and confidence risks
(00:20:16) - AI policies, guardian agents, and enforcement
(00:22:19) - Deep dive into the context layer and its fragmentation
(00:26:19) - Joe on communicating AI concepts and instilling doubt in agents
(00:31:51) - Sponsor break: Collibra as the enterprise AI control plane
(00:32:40) - Defining and evolving the semantic layer
(00:36:04) - Martin's industrial revolution analogy and guide rails
(00:39:23) - Joanne's agentic A-list framework for governance
(00:44:54) - Implementing AI governance in practice
(00:49:10) - Clarifying Collibra's AI governance offering
(00:52:35) - Trust, provenance, and lineage as governance foundation
(00:55:00) - Closing thoughts: business case, layered governance, and closing remarks

Denne episoden er hentet fra en åpen RSS-feed og er ikke publisert av Podme. Den kan derfor inneholde annonser.

Episoder(40)

SaaS, Cloud, and AI Contracts: Where Technology Leaders Lose Leverage

The 186th episode of Coffee with Digital Trailblazers focused on negotiation tactics for SaaS contracts. Key topics included competitive bidding, avoiding auto-renewals, and ensuring data ownership. M...

31 Aug 1h 2min

Shadow AI at Work: Real Risks, Practical Guardrails, Hidden Innovation

The real risks and practical guardrails around shadow AI, which goes beyond traditional shadow IT with new risks including data leakage, agentic AI poor decisions, and potential legal compliance issue...

24 Aug 1h 1min

Beyond Legacy Processes: Engineering the High-Velocity Enterprise

The session sponsored by Camunda, featuring CTO Daniel Meyer, focused on transforming legacy business processes for high-velocity enterprises in the AI era. Meyer highlighted the advantages of orchest...

17 Aug 1h 1min

Networking in the AI Era: A Guide for Digital Transformation Leaders

An expert advisory panel discussed networking in the AI era, emphasized the importance of authenticity, curiosity, and relationship-building. Panelists explored strategies for meaningful connections a...

11 Aug 1h 1min

Evaluating Technologies in the AI Era: Value, Risk, Ecosystem Evaluation

This week's "Coffee with Digital Trailblazers" episode featured industry analyst Maribel Lopez discussing the evaluation of AI technologies. Key topics included the significance of aligning AI selecti...

31 Jul 1h 2min

Does IT Leadership in the AI Era Require an IT Background Anymore?

The 180th episode of Coffee with Digital Trailblazers explored whether IT leadership in the AI era necessitates an IT background. Host Isaac Sacolick and guests discussed the importance of IT knowledg...

17 Jul 59min

Reskilling Mid-Career Leaders: What Senior Talent Needs to Stay Relevant

Our discussion focused on reskilling mid-career leaders in the AI era, featuring Dean Cantave. Participants shared experiences on adapting to AI’s impact on careers, emphasizing critical thinking, col...

13 Jul 1h