Value Driven Data Science

AI hallucinations get all the attention. But hallucinations are relatively easy to catch because the output is obviously wrong. The failure mode that should worry data scientists more is when the agent uses facts that are true to draw conclusions that are false, producing outputs that look perfectly fine. This is known as silent correctness.

In this Value Boost episode, Jia Huang joins Dr Genevieve Hayes to explore why silent correctness is the most dangerous failure mode in agentic AI systems and what data scientists can do to catch it before it causes serious harm.

You'll discover:

Guest Bio

Jia Huang is a lead research engineer at A*STAR, Singapore's Agency for Science, Technology and Research, and is the author of multiple books on AI engineering and agent design, including Designing AI Agents and RAG from First Principles. His work focuses on turning agentic AI from impressive demos into reliable, auditable, and value-producing engineering systems.

Links