Reposted by Matthew WickerAmartya Sanyal @amartyasanyal.bsky.social · 02/12/2024Two weeks remaining to apply to this position. I'll also be at NeurIPS, if you want to chat you can DM or email me. :) 042
Reposted by Matthew WickerMaike Osborne @maosbot.bsky.social · 02/12/2024AI safety is not just a topic for science fiction—AI models already entail many possible current-day hazards, including encouraging self-harm, sexual exploitation and hate speech. For a taxonomy of potential hazards, I recommend this recent paper: arxiv.org/pdf/2404.12241arxiv.org 4367