Simon Willison’s Weblog: A quote from Aaron Boodman (Oct 28, 2025) “Claude doesn’t make me much faster on the work that I am an expert on. Maybe 15-20% depending on the day. It’s the work that I don’t know how to do and would have to research. Or the grunge work I don’t even want to do. On this it is hard to even put a number on. Many of the projects I do with Claude day to day I just wouldn’t have done at all pre-Claude. Infinity% improvement in productivity on those.”
Anthropic: Piloting Claude for Excel (Oct 28, 2025) Claude for Excel can analyze complex spreadsheets, including formulas and dependencies, providing explanations with cell-level citations.
Google: How a Gemma model helped discover a new potential cancer therapy pathway (Oct 15, 2025) The model, C2S-Scale 27B, generated a novel hypothesis about cancer cellular behavior, specifically identifying silmitasertib as an interferon-conditional amplifier for antigen presentation, which was subsequently validated experimentally in living cells.
Edward Zitron: OpenAI Needs $400 Billion In The Next 12 Months (Oct 17, 2025) Perhaps OpenAI’s ambitious plans to build massive data center capacity are unrealistic and driven by market manipulation, as they lack the necessary funding, resources, and infrastructure within the proposed timelines. Time will tell.
Forbes: AI Talent Isn’t Coming To Hollywood—It’s Already Here (Oct 28, 2025) In September 2025, AI-generated entertainers Tilly Norwood, an AI actress, and Xania Monet, an AI music artist, gained mainstream commercial traction and a record deal (for Monet).
WSJ: Large Language Models Get All the Hype, but Small Models Do the Real Work (Oct 31, 2025) While large AI models grab headlines, companies are finding smaller, more specialized AI models are more effective and cost-efficient for most corporate tasks. These smaller models are often strung together in “AI factories” to automate workflows, with larger models used sparingly for complex planning and report generation.