Preamble by Preamble is an AI security and safety platform designed to control AI risk and ensure AI systems operate within defined boundaries. It supports various large language models (LLMs) including OpenAI, Anthropic, Mistral, and IBM watsonx. The platform enables organizations to curate and deploy AI policies, test them in real-world scenarios, and integrate AI models into existing workflows while maintaining security and compliance.