OpenAI CFO: July annualized revenue surpassed entire Q2

A University of Cambridge report on 332 respondents found 51% of published UK novelists believed AI would likely replace their fiction writing entirely, and 39% said generative AI had already cut their income. 85% expected future earnings to decline.

Simon Willison reran Colin Frasier's GPT-4o "sum in words" experiment locally on a DGX Spark using Qwen3.8-27B-Q4_K_M.gguf, one shot per pair with reasoning enabled. The model got 167 of 169 attempts right; with reasoning disabled he ran 30 samples per combination.

Chakra has conducted more than 500,000 interviews during its roughly six-month beta, with Snowflake, Snorkel, and Capgemini among early testers. It evaluates critical thinking and "AI fluency" — how candidates frame problems for AI and judge its output — not just final answers.

Google Research VP Yossi Matias' session covers AI reshaping biology, infrastructure, manufacturing, and science. Full three-day program on demand for $596 with subscriber discount; other sessions cover Google Quantum AI and energy systems.

Together Link plugs open models like GLM 5.3 and Kimi K3 into existing coding harnesses, claiming over 50% lower model spend. It supports Claude Code, Claude Desktop, Codex, OpenCode and Pi, with an Auto router that sends quick fixes to cheap models and hard tasks to frontier ones.

Khan Academy's Katie, a former science teacher and content creator, walks through how mastery learning is implemented on Khan Academy with AI support, aimed at building strong foundations and high expectations for every student.