VS Code to Enable Co-Authored-by Copilot Trailer by Default
Microsoft has opened PR #310226 to enable AI co-author trailers in VS Code's Git extension by default. The change flips the git.addAICoAuthor configuration default from "off" to "all". This means that whenever AI-generated code contributions are detected (likely via Copilot), VS Code will automatically append a Co-authored-by trailer to commit messages.
Key Details
- The PR modifies
extensions/git/package.jsonto set the default value ofgit.addAICoAuthorto"all". - A review from Copilot AI itself flagged a potential issue: the runtime fallback in
extensions/git/src/repository.tsstill callsconfig.get('addAICoAuthor', 'off'), which could lead to unexpected behavior in contexts where the contributed defaults aren't loaded (e.g., some tests or hosts). The suggested fix is to update the fallback to match the schema default or omit it entirely. - The PR is authored by cwebster-99 and received significant community reaction — 216 thumbs-down and 24 confused emojis at the time of this writing — indicating developer pushback against adding co-author attribution by default.
If merged, every commit made while Copilot was active will include a Co-authored-by: Copilot trailer unless the user explicitly opts out. This change is part of ongoing discussions about proper attribution for AI-assisted development.
📖 Read the full source: HN AI Agents
👀 See Also
Litigation Risks in AI Data Center Financing Structures
The AI data center buildout requires $5.2 trillion in infrastructure investment by 2030, with companies using complex financing structures like SPVs and GPU-collateralized facilities that create nine categories of litigation risk.
Google Chrome Installs 4 GB Gemini Nano AI Model Silently – No User Consent
Google Chrome has been found to silently download and install the 4 GB Gemini Nano AI model on user devices without explicit consent, sparking privacy and storage concerns.
Research shows personality affects Claude's self-correction, not Llama or Qwen
A researcher ran 23 experiments testing self-correction without guardrails across Claude, Llama, and Qwen. The main finding: personality profiles affect Claude's self-correction ability, with high directness catching all errors and low directness catching none. Llama and Qwen didn't self-correct even with identical prompts.
M5 Max vs M3 Max Inference Benchmarks for Qwen Models on oMLX
Benchmarks comparing M5 Max and M3 Max MacBook Pros running Qwen 3.5 models via oMLX v0.2.23 show M5 Max delivering 1.4-1.7x faster token generation and up to 4x faster prefill at long contexts.