AI (Sifters) FAQs

Yes. Sifters are consistently improved with the help of Sifter Trainer reports and focused research and development efforts from our Data Science and Content Development teams. When we started in 2013, our Sifters were performing at F1 levels around 90%. Now they exceed 97%. We will continue to improve the performance of our Sifters as our user base grows and our technology advances.

Yes. Just like people, our Sifters make errors. Clients and team members reportSifter errors through our Sifter Trainer, a feature of both LegalSifter ReviewPro and LegalSifter Organize. Once our Content Development team confirms the accuracy of the report, we will retrain the Sifter that made the error within 5-10 business days and release the improved Sifter back into production.

Yes. We use advanced machine learning to keep our Sifters improving.

Yes. We use advanced NLP in our Sifters. The technology, among other things, allows our Sifters to read and understand novel language that they have never seen before. The ability of Sifters to read and understand novel language is critical in contracts, as most agreements have unique phrasing.

On average, our F1 accuracy score is 95%, with our latest Sifters at 97% or higher. These F1 scores materially outpace human performance. That doesn’t mean that the Sifters are “smarter” than people, but it does mean that a collection of Sifters is more likely to accurately classify and analyze contract concepts than a human.

ReviewPro uses a combination of artificial intelligence techniques to deliver precise, reliable results:

Sifters. Sifters are contract-specific algorithms that use Machine Learning (ML) and Natural Language Processing (NLP) to identify and flag critical issues or missing terms in your contracts. Each Sifter is trained on thousands of real-world contracts, released to clients when it achieves 95+% accuracy, and continuously improved with expert feedback.

Generative AI. Generative AI automatically applies precise redlines that align exactly with your company’s established positions. We tightly control and constrain the AI, so that it operates like an attorney under strict supervision. We measure AI quality using thoroughness, accuracy, and edit fluency—consistently achieving a 95%+ accuracy rating. Together, these advanced AI technologies ensure ReviewPro delivers consistent, attorney-quality results, faster and more reliably than any other solution available.

A Sifter is a piece of software trained to read text and look for a specific concept. It reads novel language, learns from experience, and improves over time. Our Sifters use machine learning (ML) and natural language processing (NLP), types of AI, along with a number of other advanced search techniques.