Testing AI Fairly - QonfX London Fireside Chat – QonfX London 2026 Fireside Chat

FS: "Are We Testing AI Fairly? Bias in the Evaluation Process Itself"

Abstract

Most responsible AI conversations focus on bias inside the model. This panel flips the lens onto the testing process itself: the test data, eval scenarios, and grading criteria QE teams design are just as capable of encoding bias, and far less scrutinized. If the yardstick is skewed, passing tests tells you nothing. This panel gets into how quality engineering can audit its own evaluation methodology, not just the AI being evaluated.

Key Takeaways

About Panelists

Venkatesh Sriraman

Head of Software Engineering Group - UK&I at Cognizant

Venkatesh Sriraman is UK&I Head of Software Engineering at Cognizant, where he leads the Software Engineering business and helps enterprises adopt AI-driven software development. He has more than two decades of experience across the UK, Asia-Pacific and the Middle East, spanning digital engineering, quality engineering and AI. Having built and scaled Cognizant’s Quality Engineering business in the UK and Ireland, he brings a practical perspective on why trust in AI depends on strong quality management.

Sabarinath Keerthivasan

Head - QE & AI Assurance at Infosys

Sabari brings 25 years of global experience across consulting, sales, and delivery, helping clients navigate enterprise-wide transformations. In his current role, he leads Quality Engineering offerings across EMEA and drives the go-to-market strategy for AI Trust and Assurance. A strong advocate for Quality Engineering evolution, Sabari actively evangelizes AI diffusion and AI assurance strategies for frontier enterprises.

QonfX