Claude
What did we learn from it: Capybaras, undercover mode, and unreleased features...
Similar to "academic and professional research" processes, says Microsoft of defaulting to two-engine research mode.
Claude is wildly outperforming GPT and Gemini in BullshitBench. Its creator thinks some vendors may be losing touch with fundamentals.
Google's numbers suggest it could overtake OpenAI in enterprise seats this year, as Claude vows it will never go ChatGPT's way on monetising consumers.
For enterprise users, Claude's big new soul doc puts the brakes on agentic use and implies limits in using it in coding pipelines.
Claude Code users are complaining over tighter usage limits after a holiday offer reset.
The buzz is deafening. But GitHub bug reports flag warning signs over capacity buffers...
Steve Klabnik co-wrote the book on Rust. LLMs may now be good enough that he can write a better compiler in Rue, he thinks.