Airlock is a quality control solution for AI-generated text that works to help prevent the disclosure of sensitive information, such as PII and PHI, and to control the sentiment and offensiveness of your AI-generated text. Airlock analyzes your AI-generated text according to your business policies that control how sensitive information is managed. By using Airlock as an AI policy layer between your AI models and your users, you can lower the risk of your AI applications exposing sensitive information or offensive text to your users. Airlock features a simple API for applying your policies to your AI generated text, and can identify many types of PII, PHI, and other types of sensitive information. Airlock runs entirely within your cloud and does not require access to any external services outside of your VPC.