All articles tagged with #code review
Godot Foundation updates its contribution policy to require all code be human authored, banning autonomous AI agents, AI-generated code, and AI-generated text in human-to-human communication, amid reviewer shortages and a push to cultivate future maintainers.
Cursor data shows a rising share of AI-generated code changes reaching production without a separate human review in the past six months, signaling growing developer trust in AI to handle larger portions of software development. While quality metrics aren’t fully measured, AI-produced changes are surviving at higher rates, hinting that AI coding is becoming more reliable and could reduce the need for manual review, with implications for workflows and human roles.
Linux kernel developers’ b4 patch tool now features a text-based UI that supports AI agent-assisted code reviews using Claude Code, enabling an opt-in, agent-enabled review of patches within the b4 workflow. Konstantin Ryabitsev calls this the first actual agent-assisted review on patches via the new TUI, with ongoing refinements to come. The development complements other AI-prompt work in the Linux kernel space, including efforts from Meta’s Chris Mason.
Linus Torvalds publicly criticized a late and poor-quality RISC-V code contribution from a Google engineer, calling it 'garbage' and warning against future late submissions and low-quality code, emphasizing the importance of early and clean pull requests for the Linux kernel.