AISU Safety Research Talk - Dmitri Krashenninikov @Anthropic

​Hosted by AI Safety Utrecht (AISU) both Virtually (Gmeet) and in person

​📍Location: Utrecht University - Drift 23, 2.12

​How does AI safety research look in practice at a frontier lab?

​Join AI Safety Utrecht (AISU) for an online guest talk and Q&A with Dmitrii Krasheninnikov, Research Fellow in AI Safety at Anthropic. The session is aimed at students and early-career researchers with a technical background (ML, CS, or related fields) who are curious about applied AI safety research.

​🎤 Speaker

​Dmitrii Krasheninnikov
Research Fellow, AI Safety @ Anthropic
PhD in Machine Learning, University of Cambridge

​🔍 What to Expect

• ​A research-focused talk (15–20 min) on applied AI safety

• ​Examples of alignment, evaluation, and safety challenges in real systems

• ​Reflections on open problems and research directions

• ​Live Q&A with technical and career questions

​⏱️ Format

• ​Duration: ~60 minutes

• ​Format: Online talk + Q&A

• ​Audience: Students & early-career researchers (ML / CS background)

• ​Timezone: CET (Netherlands)

​This event also serves as the kick-off session for AISU’s upcoming AI Safety Fundamentals program.



Find more about our courses, collaborations and community: https://aisafetyutrecht.org/