Juro, which focuses on corporate contracting needs, has launched ‘AI Extract’ to accelerate the review of incoming third-party contracts.
The new feature will analyse contracts in relation to pre-set playbooks, as well as tag and aggregate key elements of a document. In addition, you will be able to automate approval workflows and even get translations. All of which improves the contract triage process for inhouse lawyers.
The move is part of Juro’s growing use of genAI capabilities and builds on its AI Assistant, an editor-based tool that enables users to draft, review and summarise contracts.
Richard Mabey, CEO of Juro, said: ‘Business users should be able to get contracts agreed however and wherever they want, whether that’s on their paper, your paper, Docx, PDF or natively in Juro – and it should be accelerated by the most powerful AI at the point of requirement. We’re proud to offer that experience to our customers with AI Extract.’
Is this a big deal? Well, the field of contract review has always been a key target of legal AI enthusiasts, starting with the first wave of ML/NLP tools in the early 2010s. But, generative AI capabilities have now taken things to the next level.
This in turn has given CLM and contract management providers additional abilities to wield for the clients. For example, as noted above, such as even being able to add in translation as part of the workflow above and beyond the review based on a playbook.
And on that point about playbooks, plenty of companies have tried in the past to automate this complex task with ML/NLP tools, but outputs were often not quite as good as desired. This in part was because conceptual aspects of a playbook, or where there was chained logic, were not always able to be picked up.
E.g. an automated playbook that specifies a term must be X years may spot a discrepancy, but where it may refer to language that is open to interpretation it is much harder to get right. LLMs, while also having some of their own challenges, are better at conceptual language understanding and handling the intention of a playbook, rather than having to depend on simple word matching. This then makes handling third-party paper via an automated playbook a much more effective solution and turns what many were aiming at in the past now into a reliable process.
In turn, this all leads to the broader desired outcome when it comes to deploying legal AI: improved legal efficiency – although in this case for a business.
Kyle Piper, Contracts Manager at ANC, a US sports technology company, said of the capability: ‘I’ve been able to get twice as many documents processed in the same amount of time while still maintaining a balance of AI and human review. This AI functionality feels like the next step for intuitive CLM platforms.’
So, there you go. It has perhaps taken the legal tech sector a bit longer than hoped to get to this kind of position, but now we are here. The next question is what will inhouse legal teams do with these time savings? Well, that will have to be the subject of another piece in AL.