Google DeepMind Opens AGI Institute to Broaden Debate

Base Labs, the research group spun up by Baseten earlier this year, has launched a partnership with Hugging Face and Goodfire to develop and publish methods for training and monitoring open-weight AI models. The collaboration focuses on AI safety practices for open models, addressing a gap in standardized approaches to model development and oversight. The partnership will produce publicly available methods and tools for the open-source AI community.

OpenAI is close to solving the Hodge Conjecture, a second Millennium Prize Problem, following earlier controversy over its work on the Navier-Stokes problem. The company is deliberating how to announce the solution collaboratively with the math community to avoid repeating a recent public relations incident. The timing of the announcement remains uncertain as OpenAI weighs its approach.

OpenAI research scientist Noam Brown stated that the company's top priority when training new AI models is automating AI research and development, describing recursive self-improvement as the number one goal by a wide margin. While GPT-6 Astra showed improvements across professional tasks including video game design and sheet music transcription, Brown emphasized that these capabilities are secondary to the core objective of enabling AI to improve itself. Brown, who has spent three years at OpenAI focusing on AI reasoning and autonomous agents, discussed these priorities in an interview for The Information's new AI Deep Dive series.

Researchers at Nature Machine Intelligence have demonstrated a dual-model architecture for autonomous crystal materials research using two lightweight large language models working collaboratively. The approach combines reasoning and scientific tool execution while maintaining computational efficiency and local deployability. The method achieves competitive performance without requiring expensive infrastructure, making advanced materials research more accessible.