## LLVM Adopts Human Oversight for AI-Generated Code The LLVM open-source project has determined that code contributions developed with the aid of artificial intelligence (AI) tools will be accepted, but only if supervised by a human. This decision follows an internal debate on the advisability of integrating automatically generated code. The new policy stipulates that a technical expert carefully examines the code produced by the AI before any pull request. Contributions generated exclusively by AI, without any human verification, will not be allowed. This measure aims to balance the benefits offered by AI in terms of speed and automation with the need to ensure the quality, security, and maintainability of the code. Human supervision is considered essential to identify potential errors, vulnerabilities, or compatibility issues that may escape automatic analysis. LLVM is a fundamental project in the software development tool ecosystem, used to create compilers, debuggers, and other essential tools. Its importance makes a cautious approach to the integration of new technologies such as AI crucial.