If AI safety achieves one thing before 2027, what should it be? - Lightning Talks & Mixer
A dynamic evening bringing together researchers, policymakers, engineers, and independent thinkers from across the AI safety ecosystem for two rounds of short lightning talks, followed by an informal discussion over snacks and drinks.
The evening is centred on the question:
“If AI safety achieves one thing before 2027, what should it be?”
The second half of 2026 is shaping up to be a pivotal moment for the field. AI capabilities are advancing rapidly, and many core questions and concerns remain unresolved. This event is designed to create space for bold, candid, and forward-looking perspectives on what matters most now.
Speakers will each have five minutes to present one key idea, proposal, warning, priority, or strategic direction for the field. The lightning talk format encourages concise, opinionated contributions and exposes attendees to a wide range of viewpoints across technical safety, governance, policy, coordination, and research.
Following each round of talks, attendees will have the opportunity to engage speakers in an open Q&A and continue conversations over snacks and drinks.
Speakers
Confirmed speakers include:
Dewi Erwan, Co-founder and CEO of BlueDot Impact
Jonah Wilberg, Co-lead of Life Itself Research
Jonas Sandbrink, Entrepreneur-in-Residence at Sentinel Bio
Monika Jotautaité, Technical Staff at Apollo Research
Renan Araujo, Director of Programs at the Institute for AI Policy and Strategy (IAPS)
Stefan Torges, Head of Strategic Initiatives at Forethought
Tyler John, AI Program Director at Effective Institutions Project
Doors open at 18:30, and the programme will kick off at 19:00.
Space is limited, so please register for the event if you would like to attend. We will not be able to accomodate drop-ins.