What does it mean for AI to be “aligned”?

Alignment has become a buzzword that frequently surfaces throughout the AI ecosystem. But what does it mean to align LLMs to the hundreds of millions of people who now interact with their outputs, each with different preferences for language, conversational norms, value systems, and political beliefs? This talk will explore the gnarly concept of alignment, specifically for large language models (LLMs). Using insights from the recently released PRISM Alignment Dataset, we’ll cover key questions around empirical alignment efforts, such as how to collect human feedback using various scales and signals, where to focus human labour, who to ask for feedback, and whether to target objectives of personalised or collective alignment. Through more comprehensive and representative data, PRISM highlights the challenges of aligning LLMs to meet the diverse expectations and beliefs of a global audience.

Please join the event. Everyone is welcome—it is free and you do not need to be affiliated with the university.

About Hannah Rose Kirk

Hannah Rose Kirk is a PhD student at the University of Oxford, UK. Hannah’s research centres on the role of granular and diverse human feedback for aligning large language models. Her body of published work spans computational linguistics, computer vision, ethics and sociology, addressing a broad range of issues such as AI safety, bias, fairness, and hate speech from a multidisciplinary perspective. Hannah holds degrees from the University of Oxford, the University of Cambridge and Peking University. Alongside academia, she collaborates often with industry projects at Google, OpenAI and MetaAI, and previously worked as a Data Scientist in the Public Policy Team at The Alan Turing Institute.