Cover Image for HASH AI Safety Evening: Talk by Dr Yi Jingwei (BAAI) + Lightning Talks + Networking
Cover Image for HASH AI Safety Evening: Talk by Dr Yi Jingwei (BAAI) + Lightning Talks + Networking
Avatar for HK AI Safety Socials
Monthly reading groups/social events. Also check out our speaker events here: lu.ma/hashspeakers

HASH AI Safety Evening: Talk by Dr Yi Jingwei (BAAI) + Lightning Talks + Networking

Register to See Address
Hong Kong Island, Hong Kong
Registration
Approval Required
Your registration is subject to host approval.
Welcome! To join the event, please register below.
About Event

​Join us for our upcoming AI Safety Evening on 27 November 7:00–9:30pm, at the Hong Kong AI Safety Hub @ Le Bureau, in Central.

​We're honoured to welcome Dr Yi Jingwei from the Center for Large Language Model Safety at the Beijing Academy of Artificial Intelligence (BAAI) as our special guest. She will give a presentation on her current research on Pluralistic Value Alignment: Challenges and Technologies.

Dr Yi's research focuses on responsible artificial intelligence, with particular interests in the societal impacts, safety, and reliability of large language models. She has conducted extensive research on jailbreak attacks and defenses, safety alignment, and deceptive behaviors in AI systems. Her work has been published in leading journals and conferences, including Nature Machine Intelligence, ACL, KDD, and EMNLP, and has been deployed in large-scale industrial applications such as Bing Search and the MSN News recommendation system.

​Rundown: 7:00 registration & networking · 7:30 guest talk + Q&A + lightning talks · 8:30 networking.

​Call for lightning talks: We're inviting 5-minute talks from Hong Kong's AI safety community—technical alignment, evaluations, governance, risk, safety engineering, and more. Lightning talks are the fastest way to put your work in front of Hong Kong’s growing AI safety community. In five minutes you can:

  • ​Share a research finding or work-in-progress with a technically engaged audience

  • ​Float an idea and get feedback before it becomes a full paper

  • ​Connect with collaborators across universities, labs, and industry

  • ​Practice concise science communication—a skill that matters as much in AI safety as anywhere

​Whether you’re a PhD student, faculty member, independent researcher, or working in industry, if you’re thinking seriously about AI safety research, we want to hear from you.

​Submit here: Open Google Form

​Submissions close November 22, 2026.

​Who should attend: Researchers, students, data scientists, AI engineers, policymakers. Come to present, come to listen, or come for the conversation—the networking sessions are where much of the value happens.

​Please note: Capacity is limited and registration is subject to approval. Only approved registrants will be admitted.

​Food & Drinks: Non-alcoholic drinks and light vegetarian canapés will be provided.

Location
Please register to see the exact location of this event.
Hong Kong Island, Hong Kong
Avatar for HK AI Safety Socials
Monthly reading groups/social events. Also check out our speaker events here: lu.ma/hashspeakers