Ironclad Journal icon Ironclad BLOG

The Strong Get Stronger with Contract AI

October 3, 2022 3 min read

Ask ten people the most exciting part of AI, and you’ll easily get ten different answers. That’s because AI is such a huge thing with broad implications. It will dramatically change our lives over the next decade, and we’re all anticipating different aspects. For me, the most exciting part is seeing how AI’s strengths and weaknesses will continue to complement the strengths of human intelligence.

AI is incredible. When we first started concepting Ironclad AI in 2017, I was floored by its ability to recognize complex legal clauses that a human adult would have struggled with. But what we really needed it to do was identify more straightforward things like the “counterparty name” and “agreement date.” Without that, the contract AI wasn’t smart enough to be useful. 

At first, AI utterly failed to perform simple tasks – at least, that’s how I thought about it at the time. 

It took me a while to understand that what I perceived as the AI failing on simple tasks, was actually me failing to recognize its strengths. I was using human intelligence as a measure of its ability. AI is its own unique type of intelligence, and there were already things the AI could do far better than a human. 

Contract AI doesn’t make lawyers obsolete; it makes them stronger

This is what we’re doing with Ironclad AI: bringing together machine learning and human intelligence, so that the whole is greater than the sum of its parts:

  • In Ironclad’s contract editor, “playbooks” configured by human intelligence enable Ironclad AI to flag out-of-bounds provisions and recommend alternative language.
  • In Ironclad’s Repository, structured data extracted from contracts by Ironclad AI can be combined with powerful full-text search queries crafted by human intelligence to pin-point necessary contracts.
  • And with Ironclad Insights, users can create data visualizations of in-flight contract data that Ironclad AI keeps updated in real-time.

This is also what differentiates Ironclad AI: the ease-of-use necessary to bring together AI and human intelligence. But that ease-of-use is only achievable because of the power and accuracy of the underlying AI models – and because the user experience has AI deeply embedded, rather than bolted on.

What’s next for Ironclad and contract AI 

What about the future? With AI, it’s really difficult to predict. Progress on AI comes in leaps and bounds. In seeking to understand AI, I’m very aware that I don’t know what’s next. But I’m excited to be surprised by it. One prediction is safe: the strong will continue to get stronger with things like contract AI. 

Closer to home, we’ve seen AI excel at NER (“named entity recognition,” like the “simple task” of identifying things like “counterparty name” and “agreement date”) in the fairly recent past; this will help us increase Ironclad AI’s extraction capabilities to 100 data fields and beyond this year. 

On top of that, Ironclad AI has recently started showing promise in comparing the differences between contract documents. The first place we’re seeing this is in contract negotiation and tracking contract data as it changes in real-time from version to version. We expect this to also have strong applications in analyzing large amounts of similar documents, like sales agreements that all started from the same template.

We’ve got an exciting contract AI roadmap ahead. Stay tuned for more announcements this year. In the meantime, sign up for a custom demo to see how Ironclad AI helps you work faster, streamline operations, and collaborate more effectively.

Want more content like this? Sign up for our monthly newsletter.

Book your live demo

Cai GoGwilt is CTO and Co-Founder of Ironclad. Before founding Ironclad, he was a software engineer at Palantir Technologies. He holds a B.S. and M.Eng. in Computer Science from the Massachusetts Institute of Technology.