curl maintainer: AI Rust rewrites lack 200+ bugfixes merged in 3 weeks
Daniel Stenberg, curl's maintainer, noted that AI coding agents converted 185,000 lines of C into 340,000 lines of Rust, but during that period the real curl project merged over 300 commits, including 200+ bugfixes and several vulnerability fixes, which the rewrites lack. He questioned whether these clones can keep up with ongoing improvements.
Coverage timeline
Simon Willison
During the time these agents converted 185,000 lines of C code into 340,000 lines of Rust code (I counted), we merged over 300 commits in the actual curl project. We actually improved curl during these a little more than three weeks. A little over 200 bugfixes actually. 200 bugfixes that none of those rust rewrites have. We also fixed a few vulnerabilities. In another month we have done another 200 and maybe fixed a few more vulnerabilities. Will these clones keep up? — Daniel Stenberg , Actual curl rewrites in Rust Tags: daniel-stenberg , curl , rust , security , coding-agents , c , ai , llms , generative-ai , vibe-porting
Related stories
GitHub Octoverse 2026: AI agents triple commits to 7.38B per month
At GitHub Universe, the 2026 Octoverse report showed that AI-agent-driven activity has tripled merged pull requests to 220 million per month and commits to 7.38 billion per month, with most growth in the second half of the year. GitHub has added 4 million CPU cores and 160PB of storage in 2026 to support this growth, and new repository creation has doubled to 30.3 million per month.
2 reports · 2 sources

Revolut moves ~2,000 engineers to self-hosted GLM, cutting AI coding costs by 90%
Revolut has moved roughly 2,000 engineers to self-hosted GLM models for coding, making frontier APIs like Claude Code and Codex the exception rather than the default. The company reports a ~90% reduction in AI coding inference costs and a 70% improvement in engineering lead time, attributing gains to the removal of token rationing. The shift highlights a pricing challenge for frontier model labs as open-weight models become the default choice.
2 reports · 2 sources192 points · 168 comments
News updates every hour. Subscribe in your RSS reader to get new stories automatically.
Subscribe