Last week, at the HighQ Forum in London, our new robot overlords displayed their mighty powers and declared that all human lawyers should line up and await their turn at the guillotine.
Oh wait… I’m wrong, that didn’t happen.
However, we did get a brief glimpse into the future of legal service delivery, with what could arguably be called the first true robot lawyer.
Yes, it’s a title that has been thrown around quite a bit recently. Both ROSS and KIM have been labeled robot lawyers, but ROSS is a very powerful research tool and KIM is a ‘virtual assistant’, akin to Siri for law.
Not to in any way diminish either of these technologies, if moderately pressed, I will admit to being a huge fan-boy when it comes to both of them, but I think the term robot lawyer when applied to these technologies has invited skepticism and derision from people who claim that computers simply cannot do what humans can do.
We set out to do some actual lawyering with computers.
HighQ Collaborate is a platform that allows for easy sharing and communication within firms, or between clients and firms. We may not be not the obvious choice for setting out to create a robot lawyer.
But therein lies the strength of our approach, because our robot lawyer is not a product. It’s not a creation of a single company. It’s simply a proof of concept to show what is possible when you combine resources and tools that you have at your disposal to create something that is greater than the sum of it’s parts.
This is a technique I talk about a lot, that I call bricolage.
Bricolage gives you the best of both the Buy and Build options. You are still building a custom solution to solve you particular problem. That could potentially give your firm a competitive advantage.
However, you are also using purpose built tools that are fully supported by other companies to ensure that you have the most robust solution possible. To me, bricolage is the answer to the Buy vs. Build question for law firms.
In February, HighQ announced its integration with RAVN, an AI data extraction tool that allows you to pull specific data out of unstructured documents and to move it into a structured format.
On June 9th, at our Client Forum, we also announced integration with Neota Logic, a different kind of AI that allows you to build powerful expert systems to replicate virtually any logical process that can be codified.
For the forum I was joined on stage at the British Film Institute on the south bank of the Thames, by Sjoerd Smeets from RAVN and Greg Wildisen from Neota Logic. And as a demonstration of the combined power of our three platforms, we presented a scenario:
Imagine you’re a law firm, and you are approached by a client that is considering acquiring a large number of commercial leases. They want you to help determine the value of these leases over their entire term, as well as identify any risks associated with each lease.
Now, most firms would have two options:
- Get a bunch of young lawyers, or contract lawyers, in a room and have them manually plow through the many thousands of leases, calculating the value and highlighting and risky clauses or potential concerns.
- Work with the client to identify a subset of leases to review manually, and make a number of assumptions about the rest of the leases in order to provide some likely risks they may face.
But with HighQ, RAVN, and Neota, there is a third option.
Clients will commonly upload a large set of documents into our HighQ Collaborate site. An administrator will then go through the documents, ensuring that they are appropriately filed and then notify (or set auto-notifications to notify) the appropriate lawyers that the documents are out there waiting for some attention.
In our demo last Thursday, the files were bulk uploaded and then RAVN went to work reviewing the documents.
First it identified the types of documents that were in the zip file. There were 10 commercial shopping mall leases and 5 ISDAs. As the audience watched, Sjoerd from RAVN, hit refresh and nothing happened.
He waited a second, hit refresh again, and nothing happened. He looked back at his laptop that was running the software, which I could see running, and I thought, “NOOOOO! The curse of the live demo!” I was silently screaming what an idiot I must be for trying to do this live.
But then Sjoerd hit refresh one more time, and you could see that the numbers were changing. RAVN was moving the files to the Shopping Mall Leases, and ISDA folders that we created.
Then he clicked over to iSheets, our online spreadsheet/database module, and showed how RAVN was populating the sheet with information from the uploaded documents. First one row of data showed up, refresh, four more rows, refresh, all ten. And with that Sjoerd handed the computer over to Greg from Neota.
Greg took the stage and showed the app that Neota had embedded into Collaborate. With the touch of one button marked, “Run Lease Assessment” the app performed four tasks for each lease.
It calculated the portfolio rental value from any given start date, it assessed risks associated with the calculated rental value (such as tenants right for early termination and/or assignment, special obligations on the landlord, conditions around the security deposit, etc).
Clicking through the app brings you to a valuation summary that shows the total value of the aggregated leases, as well as an aggregate Red Amber Green risk assessment of all leases. In addition, each lease is given its own valuation and risk report and the iSheet is updated with the valuation and risk report. It does all of this in seconds.
I took the stage again and did my best Steve Jobs impersonation. “That is amazing!” Except, it wasn’t hyperbole, that is actually really amazing. Several people came up to me after and said, “I’m afraid your presentation was too slick, I don’t think that everyone in the audience understood what you three just did there.”
But enough understood it. And enough can extrapolate to their own use cases and opportunities. Enough can imagine how they could then use Collaborate to share the results of the AI engines, filtering views of the iSheets and permissioning them for different audiences, the client, the practice group, the contract lawyers, and any others you could think of.
Each group seeing only the information that is relevant and important to their portion of the work at hand. Enough understood what we did on Thursday that they are beginning to talk, and they are beginning to ask whether we could make this work for their particular use case.
This robot lawyer does not replace human lawyers. It makes them faster, more efficient, more consistent, and happier.
Because this robot lawyer tells them where to focus their energies, on high risk leases, or contracts. The kinds of things that lawyers really want to do, instead of mindlessly slogging through 50 mind-numbing, perfectly normal contracts a day, hoping to find the one anomaly in a hundred contracts.
This robot lawyer doesn’t replace human lawyers. It makes them better lawyers.
Watch the full session below: