[OpenAI Releases Open-Source Model Privacy Filter for PII Detection and Redaction] Deep Tide TechFlow reports that on April 22, according to official sources, OpenAI released the open-source model Privacy Filter for detecting and redacting personally identifiable information (PII) in text. The model supports local deployment and can process long text for recognition and redaction in a single forward pass, with a maximum context length of 128,000 tokens. Privacy Filter has 1.5 billion parameters, with 50 million active parameters, and can identify sensitive information such as personal names, addresses, emails, phone numbers, URLs, dates, account numbers, passwords, and API keys. OpenAI stated that the model is released under the Apache 2.0 license on Hugging Face and GitHub and can be used in privacy protection workflows such as training, indexing, logging, and auditing.