Biography

Having grown up in Semipalatinsk, Kazakhstan – a region still impacted by Soviet nuclear testing – I developed a strong conviction that introducing emerging technologies requires careful consideration of their effects on people and society. My undergraduate and master's research at Columbia and Oxford, where I analysed social media data and observed how design could inadvertently foster polarised attitudes, further solidified this view. At Cambridge, and drawing on my experience with the development of increasingly social AI systems at Google DeepMind, I'm driven to address the current gap in understanding how AI's developing social capabilities contribute to undesirable behaviors, such as deception. This research aims to create methods for behaviourally measuring AI theory-of-mind and investigating its link to deception, ultimately contributing to a more profound understanding of AI capabilities and helping to guide the development of safer and more trustworthy AI systems.

Previous Education

Columbia University Psychology
University of Oxford Psychological Research