Ben Edelman (@edelmanben) 's Twitter Profile
Ben Edelman

@edelmanben

Thinking about how/why AI works/doesn't, and how to make it go well for us.

Currently: Technical research × policy @ U.S. AI Safety Institute

ID: 2383010678

linkhttps://www.benjaminedelman.com/ calendar_today11-03-2014 02:27:18

65 Tweet

397 Followers

48 Following

Edoardo Debenedetti (@edoardo_debe) 's Twitter Profile Photo
Edoardo Debenedetti
@edoardo_debe

10 months ago

It's great to see that AgentDojo is being used to evaluate the robustness of models that seem secure when running simple, static attacks. Amazing work from the US AISI!

Xander Davies (@alxndrdavies) 's Twitter Profile Photo
Xander Davies
@alxndrdavies

10 months ago

Cool work by Ben Edelman and the US AISI team! Happy to have helped out a bit along with Alexandra Souly :)

Cool work by <a href="/EdelmanBen/">Ben Edelman</a> and the US AISI team! Happy to have helped out a bit along with <a href="/AlexandraSouly/">Alexandra Souly</a> :)
MOSS (@moss_workshop) 's Twitter Profile Photo
MOSS
@moss_workshop

6 months ago

We are extending the deadline to May 26th 4:59pm PDT (11:59pm UTC). Thank you everyone for your interest & inquiries; we look forward to learning about your results! 🪄

MOSS (@moss_workshop) 's Twitter Profile Photo
MOSS
@moss_workshop

4 months ago

MOSS is happening this Saturday (7/19) at West Ballroom B, Vancouver Center! We are excited to have an amazing set of talks, posters, and panel discussions on the insights from and potential of small-scale analyses. Hope to see a lot of you there! 💡

MOSS is happening this Saturday (7/19) at West Ballroom B, Vancouver Center! We are excited to have an amazing set of talks, posters, and panel discussions on the insights from and potential of small-scale analyses. Hope to see a lot of you there! 💡
Xander Davies (@alxndrdavies) 's Twitter Profile Photo
Xander Davies
@alxndrdavies

2 months ago

Ben and team are extremely good—e.g. check out their cool work w OpenAI!

Ben and team are extremely good—e.g. check out their cool work w OpenAI!
Ben Edelman (@edelmanben) 's Twitter Profile Photo
Ben Edelman
@edelmanben

2 months ago

It's 2025 so you can just ask the target system to do recon on itself (Dr Andrew Fasano's innovation)