*New Paper* 🚨 Goal misgeneralization occurs when AI agents learn the wrong reward function, instead of the human's intended goal. 😇 We show that training with a minimax regret objective provably mitigates it, promoting safer and better-aligned RL policies!

a very interesting multi-criteria optimisation problem for procedural level generation in 3D games could one be inspiration for the environment design as a general sum game or a coalition game, going beyond zero-sum stuff? www.youtube.com/watch?v=Npfo... cc @michaelddennis.bsky.social

In our recent paper, we outline 15 fully risk-agnostic, process-based, evidence-seeking policy objectives. None of them limit *what* developers can do -- they just affect *reporting* and *visibility.* arxiv.org/abs/2502.096...

Very happy to announce the publication of our latest paper: A theory of appropriateness with applications to generative artificial intelligence arxiv.org/abs/2412.19010 And happy new year everyone!

Excited to reveal Genie 2, our most capable foundation world model that, given a single prompt image, can generate an endless variety of action-controllable, playable 3D worlds. Fantastic cross-team effort by the Open-Endedness Team and many other teams at Google DeepMind! 🧞

Thrilled to share Genie 2! Endless environments created by text or images, a key to open-ended/AI-Generating Algorithms. Genie 1 showed it's possible. 9 months later, Genie 2 shows jaw-dropping progress.🤯 Witness the magic of scale, again. 📈🚀 Thx to all team members @deep-mind.bsky.social!

It’s been a crazy 2 years seeing so many amazingly talented researchers bring GENerative Interactive Environments alive in Genie 1 and 2. the future is agents in generative environments

The secret to doing good research is always to be a little underemployed. You waste years by not being able to waste hours. - Amos Tversky

Welcome! @natashajaques.bsky.social. 👋 I think you may find one or two people who share your sentiment here... 😁