Prof. Alex MIHAILIDIS, Associate Vice President of International Partnerships at the University of Toronto, delivered the 19th PAIR Distinguished Lecture titled “The Future of Elder Care: Integrating Large Language Models” on 26 April 2024. The lecture attracted approximately 100 participants in person and captured an online viewing audience of over 14,100 from different countries and regions who watched the live broadcast on multiple social media platforms, including Bilibili, WeChat, Weibo and YouTube.
The lecture commenced with a welcome speech and speaker introduction by Prof. ZHENG Yongping, Director of the Research Institute for Smart Ageing (RISA), followed by an engaging presentation by Prof. Mihailidis. The content materials used in Prof. Mihailidis’s presentation were generated by ChatGPT, while he added the narrative.
In his presentation, Prof. Mihailidis first played several videos created by artificial intelligence (AI) to introduce the concept of Large Language Models (LLMs), how they can support elder care services, some drawbacks of using LLMs, and their future development. He explained that LLMs are advanced AI systems capable of understanding and generating human-like texts, as well as visual outputs and models that can respond to or interact with users. Next, Prof. Mihailidis outlined the specific benefits of using LLMs in elder care, including the support tools and resources they offer to caregivers. He then delved into the ethical considerations and challenges in LLM design, such as privacy concerns, the risk of over-dependence on technology, and barriers to technology adoption by older adults and their caregivers. To conclude, Prof. Mihailidis emphasised the role of empathetic and emotionally-intuitive AI in enhancing the quality of life for the elderly and supporting the caregiving ecosystem.
