News reporting focused on AI and machine learning, covering the companies behind these technologies, their real-world applications, and the ethical concerns they raise. This includes areas like generative AI (large language models, text-to-image and video), speech tech, and predictive analytics.
Adam Langley explores how LLMs can automate proof writing in Lean, a dependently-typed language, by building a Zstandard decompressor. He finds that LLMs can prove complex invariants in about 20 minutes, potentially making dependent types more practical for everyday software engineering. The article discusses FSE entropy coding, Lean's features, and the challenges of proof engineering.
Google Deepmind's AlphaProof Nexus framework autonomously solved nine of 353 open Erdős problems, including two unanswered for 56 years, at an inference cost of a few hundred dollars each. The system uses Gemini 3.1 Pro to generate proof steps in Lean's formal language, with compiler feedback grounding the language model's reasoning. It also tackled 44 OEIS conjectures and settled a 15-year-old algebraic geometry question.
Nine California jurors deliberate narrow questions in Elon Musk's lawsuit against OpenAI cofounders Sam Altman, Greg Brockman, and Microsoft. Issues include breach of charitable trust over Musk's donations, unjust enrichment via the for-profit arm, and Microsoft's role in aiding any breach. OpenAI counters with statute of limitations arguments tied to dates like August 5, 2021, unreasonable delay in filing until 2024, and Musk's unclean hands doctrine.
xAI added 19 natural gas turbines to its Colossus 2 data center in Southaven, Mississippi, over the past two months, bringing the total to 46 amid a lawsuit alleging Clean Air Act violations. Internal emails reveal over 500 megawatts of new capacity installed since mid-March, even after environmental groups filed suit. Regulators note emission controls but are reviewing limits on further additions.