Jack Sanderson is an undergraduate student at the University of Chicago. His research interests include open-weight model tampering, data poisoning, backdooring, and the failure modes of AI systems more broadly.