We're launching RefusalBench, a new benchmark probing how well frontier models handle biosecurity-related questions. Come for the following:

Technical talks on the challenges of evaluating frontier AI for biosecurity, followed by a social.

Biosecurity themed food & drinks provided.

We've been heads-down on biosecurity at Latch for a while now, & we're excited to finally open up & share what we've been building. This space is still small, & we think it's stronger when it's shared - so we're opening our doors to people from frontier labs , biotech, & the broader AI safety community to come hear the talks, meet the people already working in it, & see where we've landed.

We'll hear from:

[Speakers TBD]

6:00 - 7:00 Meet others. Eat + drink

7:00 - 8:00 Talks + Q&A

8:00 - 9:00 Socialize