Egor Krasheninnikov
Egor Krasheninnikov is an individual AI safety researcher who served as a Research Assistant at the Krueger AI Safety Lab (KASL) at the University of Cambridge from 2022 to 2024. His research centered on training AI systems to be helpful and interact appropriately with humans, with additional work on out-of-context reasoning in large language models and how factual knowledge is stored in model parameters. He received a £5,000 Open Philanthropy grant in 2023 to support his research in collaboration with Professor David Krueger. He has co-authored papers presented at NeurIPS 2022 and ICML 2024.
Funding Details
- Annual Budget
- -
- Monthly Burn Rate
- -
- Current Runway
- -
- Funding Goal
- -
- Funding Raised to Date
- $6,526
- Fiscal Sponsor
- -
Theory of Change
Krasheninnikov's research aimed to reduce risks from advanced AI by improving the alignment of AI systems with human intent. By developing frameworks for helpful multi-step human-AI interaction, he sought to make AI systems better at understanding and serving human needs. His work on out-of-context reasoning and source reliability in LLMs aimed to surface potential failure modes and emergent behaviors in large models, providing safety-relevant knowledge to the research community to inform safer AI development.
Grants Received
from Open Philanthropy
Projects
No linked projects.
People
No linked people.
Discussion
Sign in to join the discussion.
No comments yet. Be the first to share your thoughts.
Details
- Last Updated
- Mar 21, 2026, 7:44 PM UTC
- Created
- Mar 20, 2026, 2:34 AM UTC