Fetching from the wire…
Policy2026-07-26 · source-backed
The discussion period on GR #002 began July 24 (Debian). Matthias Geiger's would add a Section 6 to the Social Contract prohibiting any contribution "written with the use or assistance of large language models" (8 seconders, including Ian Jackson). Lucas Nussbaum's permits AI-assisted work under six conditions: tool license compatibility, contributor accountability, mandatory disclosure, prior discussion of bulk changes (9 seconders, including Stefano Zacchiroli). Ian Jackson's discourages LLMs, requires human-drafted messages to humans, and makes violations a Code of Conduct matter. Pierre-Elliott Bécue's does not endorse but accepts AI work with commit-level marking and a ban on cloud AI for confidential data (6 seconders, including Russ Allbery). Whatever wins becomes the reference policy other distros and large FOSS projects copy, so this vote matters well beyond Debian.
Each link below shares sources, entities, or timing with this story.
Project Secretary Kurt Roeckx announced the General Resolution with voting open August 15 through August 28. Options include "Ban LLM contributions from Debian via Social Contract," "Allow AI-Assisted Contributions with conditions," "Reject LLMs as far as practical" with a Cod...
Starlette 1.0.0 shipped March 22, reaching stability after 8 years at 325M downloads/month as FastAPI's foundation. Simon Willison immediately identified the problem: LLMs are trained on pre-1.0 code and will generate incompatible patterns. He built a custom Claude skill from...
GPT-5.6 Luna went to $0.20 input / $1.20 output per million tokens on July 30. That's an 80% cut. Terra dropped 20%. Luna's input now undercuts Gemini 3.1 Flash-Lite ($0.25/$1.50) and sits at one-fifth of Claude Haiku 4.5's $1 input. Simon Willison covered the announcement and...
Simon Willison highlighted Sophie Alpert's short internal policy on AI-assisted writing, whose load-bearing rule is accountability: "You must stand behind every idea and every sentence in your docs," because you can't excuse a bad passage by blaming the LLM, and doing so waste...
The decision, covered by Simon Willison, frames the ban as strategic: maintainers invest time reviewing contributions to mentor developers into trusted long-term contributors. If an LLM wrote the code, that mentorship is wasted. The wrinkle: Bun (acquired by Anthropic) runs a...
Allen Bargi's August 15 post hit 302 points arguing that AI collaboration rewards context-sharing, examples, and feedback over precise instruction (Hacker News). The pushback holds that the piece conflates management with leadership. mikeocool calls it "the most low effort ver...
MindPattern daily
One email a day at 7 AM. Sources and a take on every story. Unsubscribe anytime.