Asst prof @tudelftTBM in Data-driven/Algorithmic Systems and Safety, Justice and Sustainability. Former AI Now and Berkeley AI. Board @public_spaces.
ID: 127414067
https://bit.ly/roel-tudelft 29-03-2010 03:44:21
2,2K Tweet
2,2K Followers
1,1K Following


A holistic approach to designing, managing, and supervising public algorithmic systems. Roel Dobbe @[email protected] emphasizes the importance of considering societal factors in algorithmic decision-making, striving for equitable outcomes. Read his story here: bit.ly/3Zazmq1



SCOOP: National Institute of Standards and Technology is facing an internal crisis as staff members and scientists have threatened to resign over the anticipated appointment of Paul Christiano to a crucial position at the agency's newly-formed US AI Safety Institute, sources say. venturebeat.com/ai/nist-staffe…






I was listening to the latest Dwarkesh Patel with Leopold Aschenbrenner and I'm just kind of shocking how many times I hear the same take that is an extrapolation of trend lines guaranteeing us "AGI" by making models bigger. Honestly this feels like the prominent intelligence explosion



Want to know more about AI & climate? Join the next TU Delft AI Lunch on the unprecedented environmental impacts of AI: a transdisciplinary discussion with Benedetta Brevini, Olya Kudina, @junesallou and Roel Dobbe @[email protected] on July 2nd. Register via: tudelft.nl/en/events/2024…



NEW PAPER! 1. Small AI models often perform better than big in context. 2. Obsession w bigness has bad consequences, from climate, to power concentration, to surveillance, to research capture. This & what to do c/o Gael Varoquaux 🦋, Sasha Luccioni, PhD 🦋🌎✨🤗 and me ♥️👇 arxiv.org/abs/2409.14160