Experts explore the AI dynamics, its integration and implications

The first session of the AI Masterclass Series was successfully held via Zoom. Organised for PGDM aspirants and current students, the series aims to bridge the academic-practical divide in the ever-evolving domain of Artificial Intelligence (AI). The opening session, titled “Can We Trust AI in Business? Unpacking the Bias in Algorithms,” delved into the critical issue of bias in AI systems and its ethical implications for future business leaders.

The session commenced with a warm welcome by the moderator, Dr. Ahmad Zaman Khan. He set the tone by emphasizing the need to understand ethical and responsible AI in a data-driven world where technology significantly influences decision-making processes. He introduced the speaker for the day, Mr. Karan Gupta, a data and analytics expert from Investec. With his extensive experience at the intersection of data, financial strategy, and AI, Mr. Gupta brought a wealth of real-world knowledge to the session.

Mr. Gupta began by laying the groundwork for the discussion with a clear and concise explanation of what Artificial Intelligence is, its foundational principles, and its real-world applications. He described AI as a set of technologies designed to mimic human intelligence in performing tasks such as pattern recognition, decision-making, and forecasting. He highlighted how AI is increasingly integrated into various sectors, including finance, healthcare, marketing, and operations.

One of the most compelling aspects of the session was Mr. Gupta’s use of real-life examples to illustrate how bias can inadvertently be embedded in AI systems. From racial disparities in the US judicial system’s predictive tools to discriminatory credit scoring algorithms and biased resume screening practices, he demonstrated how AI could reinforce social inequalities if not carefully monitored.

The session also introduced key metrics used to evaluate bias in AI models, such as Independence, Separation, and Sufficiency. These were explained through simple yet impactful examples, helping students grasp how bias is measured and interpreted during model assessments.

Mr. Gupta detailed how bias can be introduced at every stage of the AI lifecycle—right from problem definition and model development to validation and deployment. He then proposed several bias mitigation strategies, including diversifying training data, auditing models across demographics, introducing fairness constraints, and ensuring stakeholder participation in model design.

The session ended with an interactive Q&A, where students asked insightful questions around the tension between profitability and ethics, and how they could prepare for careers in AI. Mr. Gupta encouraged learners to build strong analytical skills, explore ethical frameworks, and stay updated with industry tools.

Overall, the inaugural session provided a thought-provoking start to the AI Masterclass Series. With practical insights and an engaging delivery, it successfully laid a strong foundation for critical thinking and ethical awareness in AI applications for business.