🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

Convening a group of experts and building on ongoing conversations, we will address an old idea getting new attention: do we need an FDA for AI? ainowinstitute.org/general/clip-a…

Deadlines for applications extended to Monday Feb 5th!!

Crucial development as Germany has reportedly lifted its reservations on the #AIAct. FDP's digital minister, Volker Wissing, has apparently dropped its opposition. A blocking minority now seems extremely difficult. Friday's adoption is within reach. handelsblatt.com/politik/deutsc…

While the thorny issue of the scope in the Council of Europe's #AI Convention remains unsolved, a new draft has further watered down the treaty to the point it now resembles a declaration. Euractiv sheds some light on CoE's internal politics. euractiv.com/section/artifi…

❓#GenerativeAI: is it purely linguistic or is there a deeper cognitive competence? #LLMs can generate human-like text and, from scepticism to awe, they've sparked varied opinions. ➡️ CETaS' Dr Sarah asks: Can LLMs mimic human behaviour? bit.ly/4br5cWx #AI

If you believe in synergies between data protection law, data privacy law, and future AI law, then this EDPS - European Data Protection Supervisor's blog is for you: lnkd.in/ek8sUTXd.

Hype around AI in medicine often ignores two key risks: - patterns of automation contribute to centralization of power - medical knowledge is limited by the systemic refusal to trust patient expertise 1/ my new post: rachel.fast.ai/posts/2024-02-…

Interested in health and human rights? Join our Law for Health and Life workshop at the @EAHLaw conference in Warsaw 18-20 September and submit an abstract on the influence of the ECtHR on health law before 10 April 2024!🐠 More info: lchl.uva.nl/content/news/2…

🚨 We just released the results of a survey that SHOCKED me. AI is already in clinic. 🚨 We asked dermatologists -- are you using large language models in your clinical care? And then we explored what they were using LLMs for. Journal of Investigative Dermatology jidonline.org/article/S0022-…

Research alert! "How AI challenges the medical device regulation: patient safety, benefits, and intended uses" from Dr Daria Onitiu, Prof Sandra Wachter [email protected] and Prof Brent Mittelstadt @bmittelstadt.bsky.social all Oxford Internet Institute has published! Read the paper in full here: academic.oup.com/jlb/advance-ar… 1/2

🆕 ICO statement in response to Meta's plans to train #genAI with user data. “We are pleased that Meta has [...] responded to our request to pause and review plans to use Facebook and Instagram user data to train generative AI." Read the full statement: ico.org.uk/about-the-ico/…

Just came across a new paper from Xiangyu Qi, Yangsibo Huang, Peter Henderson, & Prateek Mittal on differences between safety & security in AI. Would recommend § 2 as required reading for AI governance researchers! arxiv.org/abs/2405.19524

Does more compute equate with greater risk? What is our track record at predicting what risks emerge with scale? I don't have much time anymore to write, but this felt important enough to write down some thoughts on. arxiv.org/pdf/2407.05694…

New paper and fairness toolkit alert! Governance of Emerging Technologies is delighted to release OxonFair, an open source AI/ML toolkit to enforce fairness. It includes several improvements over other open-source fairness toolkits. Paper: papers.ssrn.com/sol3/papers.cf… GitHub: github.com/oxfordinternet…

🚨New paper and fairness toolkit alert🚨 Announcing OxonFair: A Flexible Toolkit for Algorithmic Fairness w/Zihao Fu, Sandra Wachter [email protected], Brent Mittelstadt @bmittelstadt.bsky.social and Chris Russell toolkit -github.com/oxfordinternet… paper - papers.ssrn.com/sol3/papers.cf…

As we mark National Postdoc Appreciation Week, we’re celebrating the work of Daria Onitiu Oxford Internet Institute. Dr Onitiu is a postdoc researcher on the Trustworthiness Auditing in AI project and works with researchers in the Governance of Emerging Technologies programme. Find out more: bit.ly/4d8ai9P.

Join us in Oxford St Anne's College on November 12th for a Workshop exploring Trustworthy AI & Healthcare. Exciting lineup of speakers from industry and academia. Registration is free but slots are limited! More details here - oxaihealth.github.io

In medical settings, researchers & institutional users often cite medical examination benchmark results as a proxy for AI model performance across a range of medical tasks. This is a woefully misguided practice - as with actual doctors, test performance != clinical competence!

DeepSeek forces a rethink of the compute & energy costs of AI models. In a new paper with Sasha Luccioni, PhD 🦋🌎✨🤗 @strubell, we look at the full environmental impacts of AI – both direct and indirect – and what 𝐉e𝐯𝐨𝐧𝐬 𝐏𝐚𝐫𝐚𝐝𝐨𝐱 means for AI and climate. A thread: 🧵

Tensions between the AI Act and medical devices rules are emerging after the discussions around a Q&A document reached a halt last week due to significant opposition. Stakeholders consider the text to be descriptive and failing to address key concerns. mlex.com/mlex/articles/…