Desiree Cho
Interests
AI alignment, technical AI safety.
Biography
Desiree is a DPhil (PhD) student in Computer Science and a Doctoral Scholar at the Institute for Ethics in AI. She is supervised by Professor Sir Nigel Shadbolt, Chair of AI@Oxford Research. She is interested in AI safety and alignment. Her current work focuses on alignment mid-training and training diverse reward model ensembles. Her team won first place in Research Impact Oxford for their project on the latter.
Previously, Desiree completed her MSc at the University of Oxford, achieving Distinction in both her degree and thesis. In her thesis, she conducted a behavioural audit of a major social media platform: she built an original dataset of recommendation trajectories and developed an NLP pipeline to quantify recommendation drift. She has also authored user studies at leading human-centred computing venues, including the ACM CHI Conference.