If you find a technical inaccuracy that could lead readers to build insecure AI systems, report it privately rather than opening a public issue.

Email: security@system-design-for-ai.example

Include: which chapter, the specific concern, what the correct guidance should be, and supporting sources.

• A chapter recommending disabling guardrails without explaining tradeoffs
• A code sample with a known vulnerability (prompt injection, SQL injection)
• A factually wrong claim about a model's safety properties
• A recommendation that could lead to data leakage in multi-tenant systems

• Typo fixes (open a PR)
• Style complaints (open an issue)
• Pedagogical disagreements (open a discussion)