Automate your workflows even further with AI-assisted actions:
- Extract & review contract clauses for automated reviews,
- Retrieve essential data from contracts,
- Categorize and summarize documents.
Once the AI Connector has been activated, new actions become available in the module editor for all authors within this application.
Good to know
We have various options in the US, UK, and in the EU to comply with your data processing requirements, notably GDPR.
No. Your data is never used for training purposes.
BRYTER does not use your Customer Data or permit others to use your Customer Data to train the machine learning models used to provide the AI Connector.
No. Your data is fully separate from any other customer environment.
Currently, our default document support extends to 50 pages during trial, with the option to expand this to 200 pages. Both Word and PDF documents are supported. However, scanned PDFs are not compatible and must undergo OCR (Optical character recognition) before use.
BRYTER currently utilizes large language models (LLMs) provided by OpenAI and Anthropic. We continuously evaluate LLM providers and their models to provide the highest quality experience to our users while ensuring our privacy and security standards are enforced. Learn more about data, privacy, and security architecture for BRYTER.
Available AI actions
AI categorize text/ file
Determines the document type (e.g., NDA, DPA, etc.) automatically, which you can then use to direct the contract to the appropriate process or person. Use also text as an input to categorize.
Input
A document in a .pdf or .docx format or a text input.
Parameters
Configure up to five categories by default for each action, with the option to dynamically add up to 20 categories when necessary. For each category action, specify:
- A category name that will be used in your module;
- A category description that will enhance the accuracy of the LLM model to return precise results.
Categories left blank are permissible and will not be considered.
Output
Category of the input document (text).
AI summarize text/ file
Provides an overview of a (lengthy) document.
Input
A document in a .pdf or .docx format or a text input.
Parameters
Specify the number of sentences for the summary (defaults to five sentences if left empty).
Output
Summary with the defined number of sentences (text).
AI extract clause from text/ file
Checks whether a specific clause is included within the contract and extracts it.
Input
A document in a .pdf or .docx format or a text input (containing the clause).
Parameters
Include a clause description to help the LLM model to accurately extract the required clause.
Output
Text with the extracted clause. If the clause is not found, the content will be empty allowing to build some conditional logic based on this output.
AI clause compliance
Verifies if a clause adheres to a set of predefined rules.
Input
Clause in a text format, which can be linked to a clause extract action.
Parameters
Configure up to four compliance rules per clause, with at least one required to run the action.
Outputs
For each rule, two outputs are generated:
- Result: Indicates yes/no enabling you to build conditional logic.
- Analysis: Provides additional context on the result, supporting in report generation.
AI general prompt with text/ file input
Lets you prompt the LLM with any question or format you like, catering to specific needs beyond the purpose-built AI actions. These are the most versatile AI actions that unlock various use cases, including document comparison and policy context.
Input
- Text or file input prompting the LLM model from a user perspective;
- Text or file input providing additional context to the LLM, from a machine-to-machine perspective.
Parameters
Reference a text or file variable or add any type of text: a prompt, contextual information, examples, questions, lists. Optionally, add how you would like the assistant to respond.
Output
LLM answer (text): Response provided by the LLM model.
AI extract text from scanned PDF
Reads, extracts, and returns the content of a scanned PDF as text.
Input
A document in a .pdf format.
Output
Extracted text from the scanned PDF document.