OpenAI launches new privacy filter to mask personal data

In a blog article on Tuesday, OpenAI announced that it has launched a new Privacy Filter, an open-weight model designed to detect and redact personal information in unstructured text. This release aims to support safe AI development by providing practical infrastructure for high-throughput privacy workflows. Unlike traditional tools that rely on deterministic rules and pattern matching, this model utilises context-aware language understanding to distinguish between public information and personal data belonging to individuals.

The model is small enough to run locally on a developer’s machine, allowing data to be masked or redacted without leaving the device. This reduces exposure risks by removing the need to send unfiltered data to external servers for de-identification. Despite its small size, the tool is capable of processing long inputs efficiently in a single pass.

Training Announcement: The BCS Foundation Certificate in AI examines the challenges and risks associated with AI projects, such as those related to privacy, transparency and potential biases in algorithms that could lead to unintended consequences. Explore the role of data, effective risk management strategies, compliance requirements, and ongoing governance of the AI lifecycle and become a certified AI Governance professional. Find out more.