Victoria Krakovna
Source-listed: Research Scientist in AI alignment · Google DeepMind
Record snapshot: Sep 3, 2026. Coverage and affiliations may be incomplete or historical.
Selected work
5- Realistic honeypot evaluations for scheming propensitySaved credit: First author · 2026Explore people connected to this work →
- Evaluating Frontier Models for Stealth and Situational AwarenessSaved credit: Contributor · 2025Explore people connected to this work →
- Power-seeking can be probable and predictive for trained agentsSaved credit: First author · 2023Explore people connected to this work →
- Goal Misgeneralization: Why Correct Specifications Aren't Enough For Correct GoalsSaved credit: Contributor · 2022Explore people connected to this work →
- Gemini: A Family of Highly Capable Multimodal ModelsSaved credit: Contributor · 2023Explore people connected to this work →