Aleksandr Bowkis
Aleksandr is a research scientist working on AI alignment at the AI Security Institute. Previously he did a PhD in cosmology, focussing on the reconstructed CMB lensing bispectrum, at Gonville & Caius College, University of Cambridge.
Education
- PhD in Astrophysics (Cosmology), University of Cambridge — awaiting viva (submitted 2025)
- MMath in Mathematics, University of Cambridge (Distinction, 2019)
- BA in Natural Sciences (Theoretical Physics), University of Cambridge (First Class, 2018)
Publications
- Automated alignment is harder than you think (2026) — arXiv
- The reconstructed CMB lensing bispectrum (2023) — arXiv · INSPIRE-HEP