Celebrate the 200th anniversary of the Christmas Lectures with us! Former lecturers will explain how their fields have changed since their series, looking ahead to the future of research and recreating the best bits. Book now 👇 rigb.org/whats-on/disco…

Funding opportunities for promoting AI research, supported by Artificial Intelligence journal: aij.ijcai.org/wp-content/upl… **Deadline for applications: July 15th, 2025.**

TONIGHT'S LIVESTREAM 💻 Join computer scientist and Nobel laureate Geoffrey Hinton as he examines the similarities and differences between artificial and biological intelligence, following his decades of ground-breaking work. Book now 👇 rigb.org/whats-on/disco…

After so much discussion and speculation, AI is starting to take jobs. theguardian.com/technology/202…

Insilico Medicine has just shown that its AI-designed pill ― called rentosertib ― can slow or even slightly reverse lung-function loss in patients with idiopathic pulmonary fibrosis (IPF), a lethal disease. This may be the strongest clinical proof-point yet that LLM-guided

Back at Cheltenham Science Festival. Cheltenham Festivals Can AI save the NHS? Come along to find out!

2 x Postdoc positions at Oxford: Rethinking multi-agent systems in the era of LLMs. Please repost/pass on to any potential candidates! Closing date for applications: 12 noon on 4th July 2025. cs.ox.ac.uk/news/2449-full…

The future starts to arrive in London... bbc.co.uk/news/articles/…

logic

Cyberspace has caused harm, but not the global disasters many feared. Will that hold as AI advances? Blavatnik School of Government’s Prof Ciaran Martin, ex-head of the UK’s Cyber Security Centre, shares his thoughts ⬇️

Google’s Gemini 2.5 paper has 3295 authors arxiv.org/abs/2507.06261

Things are going to get very confusing… theguardian.com/technology/202…

In the last month there have been at least 3 (nonsense) papers on the Hodge conjecture posted to arXiv that I am 99% sure are LLM-generated, complete with hallucinated citations, etc. Very obnoxious to pollute the commons this way.

Today, we at OpenAI achieved a milestone that many considered years away: gold medal-level performance on the 2025 IMO with a general reasoning LLM—under the same time limits as humans, without tools. As remarkable as that sounds, it’s even more significant than the headline 🧵

You asked about how my predictions are holding up. My last slide at Dartmouth '06 was 4 immediate challenges: 1. Visual object recognition capabilities of a 2 year old child, 2. Manual dexterity of a 6 yr old, 3. Language capabilities of a 4 yr old, 4. social sophistication of a

📢 Only 1⃣ week left to apply for the role of Home Office Chief Scientific Adviser! Check out this exciting opportunity to lead on science advice for the department and apply here before midday on 28 July: gov.uk/government/new…

Do you have a PhD (or equivalent) or will have one in the coming months (i.e. 2-3 months away from graduating)? Do you want to help build open-ended agents that help humans do humans things better, rather than replace them? We're hiring 1-2 Research Scientists! Check the 🧵👇

An advanced version of Gemini with Deep Think has officially achieved gold medal-level performance at the International Mathematical Olympiad. 🥇 It solved 5️⃣ out of 6️⃣ exceptionally difficult problems, involving algebra, combinatorics, geometry and number theory. Here’s how 🧵

Rethinking multi-agent systems in the era of LLMs - 1 day workshop at Uni of Oxford - 16 Sept 2025 - Call for Expressions of Interest. (**Deadline 8 August**) sites.google.com/view/rethinkin…

Great to see Hertford College, Oxford make the wish list in “My Oxford Year”. We’re a long way below “full English breakfast”, but fair enough TBH.