Sign In

Victoria Krakovna

Senior Research Scientist at DeepMind

Victoria Krakovna is a senior research scientist at Google DeepMind, where she focuses on AI alignment.12 Her work aims to ensure that advanced AI systems behave in ways that align with human intentions and do not knowingly act against our interests.1

Professional Background

Krakovna's research at DeepMind involves:

  • Investigating how to set good incentives for advanced AI systems2
  • Exploring topics such as specification gaming, goal misgeneralization, and avoiding side effects1

She has made significant contributions to the field of AI safety, with her work being cited over 2,400 times according to Google Scholar.3

Education and Achievements

  • PhD in statistics and machine learning from Harvard University, focusing on building interpretable models12
  • Accomplished mathematician, winning a silver medal at the International Mathematical Olympiad and the Elizabeth Lowell Putnam prize2

Other Roles and Interests

  • Co-founder of the Future of Life Institute, a non-profit organization working to mitigate technological risks to humanity12
  • Member of the Board of Directors at the Future of Life Institute2
  • In her spare time, she enjoys playing with her children and spending time in nature1

Views on AI Risk

In a recent interview, Krakovna expressed her personal view that there is approximately a 10% chance of AI causing human extinction over the next 15 years.4 Her central threat model for AI risk is misaligned power-seeking.4

Victoria Krakovna maintains an active online presence, with profiles on various platforms including LinkedIn (username: vkrakovna), Twitter, Google Scholar, GitHub, and the Alignment Forum.1

Highlights

Nov 18 · twitter

Frontier Safety Framework report for Gemini 3 Pro, presenting risk assessments and evaluation results in CBRN, Cybersecurity, Harmful Manipulation, Machine Learning R&D and Misalignment domains. https://t.co/f1gLLf69np

Related Questions

What is AI alignment, and why is it important?
How does specification gaming impact AI development?
What are some of Victoria Krakovna's notable research papers?
How does the Future of Life Institute address AI risks?
What are Victoria Krakovna's views on the potential risks of AI?
Victoria Krakovna
Victoria Krakovna, photo 1
Victoria Krakovna, photo 2
Add to my network

Experience

Senior Research Scientist at DeepMind, October 2020 - Present
- Research Scientist at DeepMind, November 2016 - October 2020 - Co-founder at Future of Life Institute, March 2014 - Present - Software Engineering Intern at Google, May 2015 - August 2015 - Teaching at Harvard University, September 2012 - December 2013

Education

- PhD in Statistics, Harvard University, 2011 - 2016 - MS in Statistics, University of Toronto, 2010 - 2011 - Honors BS in Statistics and Mathematics, University of Toronto, 2006 - 2010

Location

Camden Town, England, United Kingdom