Amir Konigsberg, Ph.D.

I'm interested in what AI systems can actually do, how we can tell when to trust them, and what happens to human thinking when we begin to rely on them. My research draws on artificial intelligence, cognitive science, and epistemology to investigate these questions.

My current work focuses on several related areas:

• AI capabilities and evaluation: What do AI benchmarks actually tell us about intelligence? How can we distinguish genuine capabilities from successful performance on particular tasks?
• Epistemic robustness and AI safety: How do AI systems respond to conflicting evidence, persuasion, and adversarial pressure? How can we evaluate and improve their reliability?
• Human–AI cognition: How does relying on AI for reasoning, understanding, and decision-making affect human judgment and agency?
• Autonomous and multi-agent systems: What happens when AI systems interact with one another, exchange information, and make decisions with greater autonomy?

Alongside my research, I've spent much of the past two decades building and leading technology companies. My work has included semantic search, natural language understanding, generative AI, autonomous mobility, information integrity, and neurotechnology. I've founded, led, and existed several companies, held senior roles at Google and General Motors, and am an inventor on 18 US patents.

I'm a Research Fellow at the University of Oxford's Institute for Ethics in AI and an Adjunct Professor at Tel Aviv University. I hold a PhD in Rationality and Interactive Decision Making, and previously held research fellowships at Princeton University and the Max Planck Institute.

Outside of work, I enjoy spending time with my family, long walks in nature, painting, fermenting, thinking, and developing new ideas.

More → About.

AI Ethics