Discovering Bias in Large Language Models (LLMs)
- This session examines how bias originates and manifests in large language models.
- It is designed for technical professionals, researchers, product leaders, and policymakers aiming to understand and mitigate bias in generative AI.
- Attendees will learn about detection methods and strategies for deploying fair and ethical AI systems.
About this event
Responsible Tech Forum – Month 4 Event
ACM Distinguished Speaker Series
Date: Friday\, September 25 \| Time: 12:00 PM – 1:00 PM
Format: Distinguished Lecture + Interactive Discussion
Discovering Bias in Large Language Models (LLMs)
Speaker: Mehdi Bahrami – ACM Distinguished Speaker, Santa Clara, USA
Overview
As large language models (LLMs) become increasingly embedded in enterprise systems, healthcare, education, and everyday digital tools, it is essential to understand not only their capabilities but also their limitations and risks. While these models enable powerful automation, decision support, and productivity gains, they can also reflect and amplify societal biases, generate misleading information, and introduce unintended harms.
This ACM Distinguished Lecture will explore how bias emerges in LLMs and why it matters for developers, researchers, business leaders, and policymakers. The session will examine both the technical foundations and societal implications of bias in generative AI systems.
Participants will gain a deeper understanding of how bias manifests, how it can be detected, and what practical strategies exist to mitigate it. Through real-world examples and research-backed approaches, this lecture will advance the conversation on building more trustworthy, fair, and accountable AI systems.
What You Will Learn
What bias in large language models is and how it originates
Common forms of bias, stereotypes, hallucinations, and misinformation in LLM outputs
Current methods and tools for detecting bias in generative AI systems
Emerging mitigation strategies to improve fairness and accountability
Practical considerations for deploying responsible and ethical AI in real-world applications
Why This Matters
As organizations rapidly adopt generative and Agentic AI, ensuring fairness, transparency, and responsible deployment is essential. This session is designed to equip technical professionals, product leaders, researchers, and policymakers with the