"Addressing the risks of AI will require a mix of voluntary actions taken by companies (and private third-party actors) and actions taken by governments that bind everyone." https://www.darioamodei.com/

"Vibe coding actually works. It creates robust, complex systems that work. You can tell yourself (as I did) that it can’t possibly do that, but you are wrong. You can then tell yourself (as I did) that it’s good as a kind of alternative search engine for coding problems, but not much else. You are also wrong about that." stephenramsay.net

Cormac McCarthy notes on writing for science: gwern.net

UCLA digitized Hearst newsreels from 1929 to 1967 newsreels.net Teen-agers in their bedrooms, before the age of selfies: newyorker.com

Better late than never — iPad native Instagram app available instagram.com White paper: Why Language Models Hallucinate openai.com

Not new, but worth not missing: "Lawrence Livermore National Laboratory expands Claude for Enterprise use to empower scientists and researchers" anthropic.com

"The gpt-oss-20b model delivers similar results to OpenAI o3‑mini on common benchmarks and can run on edge devices with just 16 GB of memory, making it ideal for on-device use cases, local inference, or rapid iteration without costly infrastructure" openai.com

Another in a long line of services killed off by Google. theverge.com

bentoml.com

decrypt.co

https://doi.org/

https://www.tandfonline.com/

https://www.macrumors.com/