The ROI Rules of AI: Sirion's Legal AI (Sponsored Content)

Edward Adams (0:02)

Since you're a subscriber to this Bloomberg podcast, we thought you'd be interested in a new four-episode sponsored podcast called The ROI Rules of AI. Produced by IBM and Bloomberg Media Studios, it explores how business leaders are thinking about the return on investment of artificial intelligence projects. You can subscribe wherever you listen to your favorite podcasts. Here's a recent episode. If your company is like most, after a contract is negotiated and signed, congratulations, it gets filed away in a lawyer's office. Not to be consulted again until it has run its course and needs to be renewed.

But what if your contracts could be a roadmap to running your business? The first step in managing your relationships with your suppliers and clients. That's the function of Sirion, a contract life cycle management software for in-house legal departments. It's recently been upgraded with generative AI functionality, designed to make it something the whole company can use, not just its legal staff.

Gordon Thompson (1:12)

It gives you real-time information on how to run your business, and gives you the insights you need to pivot so that you can make the best decisions with the best data at the right time.

Edward Adams (1:22)

That's Gordon Thompson, Sirion's Executive Vice President of Presales and Business Strategy. He and his colleagues are planning a future where contracts are integrated into a company's day-to-day operations. From IBM and Bloomberg Media Studios, this is The ROI Rules of AI, and I'm your host, Edward Adams. On this podcast, we're exploring how companies of all sizes are using AI to remake their operations, increasing their return on investment and that of their customers.

Sirion has been in business for a decade. It enables companies to digitize their universe of contracts into a single database, extracting metadata like specific clauses and obligations to make them searchable. It also helps lawyers draft new contracts, making sure that the terms limit their company's risk. And it enables General Counsel's Office to measure the company's performance against those contracts. Unlike a lot of companies, Sirion has a long history of working with artificial intelligence.

Gordon Thompson (2:32)

From our inception, we were using natural language processing. As the technology evolved over the years, we started using machine learning. We've built over 600 small and medium sized language models that are purpose built to extract very specific pieces of data out of contracts at a very high precision rate.

Edward Adams (2:53)

But when it wanted to improve its contract review function and turn contracts into a device to run a business, it needed to employ generative AI. It had to move from simply copying and pasting an approved clause in place of a riskier one, to rewriting the proposed contract, as a lawyer might do. That meant using a large language model, or LLM. The problem is...

Gordon Thompson (3:20)

LLMs are expensive to build and maintain, so we wanted to create a solution that was scalable, that was cost effective.

Edward Adams (3:34)

To understand Sirion's solution, it helps to first understand what contract lawyers actually do. Lawyers call reviewing a contract redlining. And like a lot of things in the world of law, it's got some history to it. For those of the audience who are not lawyers, what does redlining a document mean?

Gordon Thompson (3:54)

So redlining got its name from when we used to do paper contracts, right? So you would send a paper contract over to your counterparty. They would literally take a red pen or red pencil, and they would review it and strike through it and make changes. That name has really persisted into the digital age. Ultimately, redlining, it boils down to making sure that you identify issues, risk, compliance, terms and conditions that meet your ultimate goals as an organization.

Edward Adams (4:21)

When lawyers redline by hand, it can be a slow process.

Gordon Thompson (4:26)

On average, it takes about 13 minutes to review a page of a contract. Some of these contracts can be several hundred pages long. When you automate it using generative AI, that goes from days, weeks down to minutes and seconds.

Edward Adams (4:39)

To speed up the process, Sirion used an open source LLM and a series of proprietary playbooks, which spell out a company's default positions on a variety of legal issues. Can you give me an example of what you mean by a playbook?

Gordon Thompson (4:56)

Your preferred position might be, I want payment in 30 days. The counterparty says that payment terms might be 60 days. So you want to be able to identify that very quickly and make sure that you put in your preferred language.

Edward Adams (5:08)

Sirion can redline even some of the longest contracts in a matter of minutes. And while human lawyers still review the software's changes, companies have found they can go from first draft to final agreement 30% to 80% faster, Thompson says. Once those contracts are signed, it took another form of AI to open up a company's database of contracts to the wider organization. IBM's conversational search technology allows designated people in a company's lines of business to ask questions of the Sirion database. And they get answers in plain English.