I’m building a new team at @GoogleDeepMind to work on Open-Ended Discovery! We’re looking for strong Research Scientists and Research Engineers to help us push the frontier of autonomously discovering novel artifacts such as new knowledge, capabilities, or algorithms, in an open-ended self-improving loop. We aim to work on ambitious research projects in a fast-paced manner. If this sounds appealing to you, apply using the link below by Friday, August 1st EOD: job-boards.greenhouse.io/dee…

Thank you for organizing another wonderful @raais edition! Had a great time speaking about a potential path towards Superhuman Scientific Discovery and chatting with everyone!

How can agents get better at algorithm discovery? Meta-meta-learning is one answer, aka improving the agents themselves at inventing generalizable algorithms. DiscoBench provides a way to procedurally generate algorithm discovery tasks at scale, which can be used for meta-meta-learning. Kudos to @AlexDGoldie and team for the release!

ARC-AGI 3 is about to drop soon. Turns out games were a great benchmark for intelligence all along? 👀 While we wait, BALROG has got you covered. Here are some new exciting results for Gemini 3 Pro, Gemini 3.1 Pro, and Claude 4.5 Opus 🔥

Gemini 3.1 Pro has landed! Amazing performance / capabilities across the board. Beyond SOTA, the best are all the things that evals can't measure. E.g. SVG has gotten so much better (see 🧵) blog.google/innovation-and-a…

🧬 New paper from my internship at @GoogleDeepMind We introduce Persona Generators: functions that generate diverse synthetic populations for arbitrary contexts. We use AlphaEvolve to optimize the generator code, hill-climbing on diversity metrics — not just likelihood — counteracting the mode-seeking behavior of LLM sampling for agent-based simulations. 🧵👇1/

Genie 3 can generate infinite worlds from just a single text or image prompt, and what's more, they are rich, interactive, and endlessly remixable. Project Genie: Available now for US Gemini Ultra subscribers. Enjoy! blog.google/innovation-and-a…

This is an outstanding PhD-opportunity if you are interested in open-endedness and video games! Check it out!

An incredible PhD opportunity! 🚀 Working with @robertarail, @_rockt, and @borruell is an absolute dream. If you are applying this cycle, put this at the top of your list!

Great opportunity to do a PhD on Open-Ended Narrative Generation with @robertarail, @borruell, and myself in collaboration between @UCL_DARK and @iconicgamesio!

We are looking for strong PhD candidates in AI to help us unlock a new universe of open-ended experiences in entertainment. This is a unique experience to work closer with the three of us and the wider teams at DARK and Iconic Interactive Research!

📢 New PhD Position 📢 We (@_rockt, @borruell, and I) are looking for a PhD student to work at the intersection of open-endedness and game design. The student will be part of the @UCL_DARK lab and funded by @iconicgamesio and UCL. See this doc for a more detailed description of the research direction and candidate expectations: docs.google.com/document/d/1… To apply, please complete this form by January 15: docs.google.com/forms/d/16JG…

A recipe for endless self-improvement toward Embodied AGI: 🧬🤖 1️⃣ Genie 3 generates the world. 2️⃣ Gemini (Task Setter) makes a task. 3️⃣ SIMA 2 acts. 4️⃣ Gemini (Reward Model) scores the video. The agent continuously generates its own experience and learns from it! [1/N]

2026 will be the year of open-endedness!

It was a great experience talking at the @SEAWorkshop at #NeurIPS. This was my first workshop talk and it’s all thanks to the organizers @lawhy_X @guohao_li, @robertarail for recommending me and @AlexDGoldie for leading DiscoBench.

🪩 DiscoBench rethinks how we evaluate AI research agents. Modular codebases so your agent can focus on the most important discoveries, meta-train/meta-test splits, and a growing set of tasks spanning RL, vision, unlearning, and more. I will be talking about DiscoBench at the SEA workshop on Dec 7th from 10:30 AM. Find me at NeurIPS if you want to chat AI for scientific discovery! 🔬

🪩 If you’re at NeurIPS, get down to Deepak’s talk on Saturday to learn all about DiscoBench!

DiscoBench makes it easy to evaluate autonomous discovery of generalizable algorithms by providing meta-train/meta-test splits for a number of diverse open research problems. Kudos to @AlexDGoldie and team for bringing this to life.

Can frontier language model agents replicate astrophysics research papers? Clearly not yet -- but models are slowly getting better! Excited to finally put out ReplicationBench, the work of an awesome team of astrophysicists from across Stanford's KIPAC, SLAC, and C4DU.

We’ve been intensely cooking Gemini 3 for a while now, and we’re so excited and proud to share the results with you all. Of course it tops the leaderboards, including @arena, HLE, GPQA etc, but beyond the benchmarks it’s been by far my favourite model to use for its style and depth, and what it can do to help with everyday tasks.

The secret behind Gemini 3? Simple: Improving pre-training & post-training 🤯 Pre-training: Contra the popular belief that scaling is over—which we discussed in our NeurIPS '25 talk with @ilyasut and @quocleix—the team delivered a drastic jump. The delta between 2.5 and 3.0 is as big as we've ever seen. No walls in sight! Post-training: Still a total greenfield. There's lots of room for algorithmic progress and improvement, and 3.0 hasn't been an exception, thanks to our stellar team. Congratulations to the whole team 💙💙💙