EncypherAI

EncypherAI is a Python package that helps embed and extract metadata in text using Unicode variation selectors. This tool is handy for invisibly encoding metadata, like model information, timestamps, and custom data, into text generated by AI models.

Benefits

EncypherAI has several key advantages. It lets you track where text comes from, so you know which AI model created it. You can also check timestamps to see when the text was made. Plus, you can embed any custom metadata you need. The tool works with both streaming and nonstreaming LLM outputs. It also has ready to use integrations with popular LLM providers like OpenAI, Google Gemini, Anthropic Claude, and more.

Use Cases

EncypherAI is perfect for anyone who needs to embed metadata into text without changing its readability. This includes developers, researchers, and businesses that work with AI generated content. The tool''s ability to track where text comes from and check timestamps makes it great for ensuring the authenticity and integrity of AI generated text.

Additional Information

EncypherAI is open source under the AGPL 3.0 license. This means you can use, modify, and share the software freely as long as you disclose the source code and any changes under the same license. For organizations that want to use EncypherAI in proprietary applications without sharing the source code, a commercial licensing option is available. This option includes benefits like proprietary integration, legal certainty, and professional support.

The package includes a comprehensive command line interface and follows PEP 8 style guidelines with Black as the code formatter. Pre commit hooks are used to automate code formatting and quality checks. EncypherAI welcomes contributions from the community, and detailed guidelines can be found in the CONTRIBUTING.md file.

For more information, you can look at the documentation for detailed integration examples and code snippets. Or try EncypherAI directly in your browser with an interactive Google Colab notebook. The notebook shows core features including metadata embedding, extraction, HMAC verification, and tampering detection.

For commercial licensing inquiries, please contact [email protected].