Please enjoy this blog authored by BIll Bice, CEO, nQ Zebraworks.
It’s been the “The Year of GenAI”. Generative AI (“GenAI”) burst on the scene in late 2022 with the public release of ChatGPT, the fastest-growing application ever. It exposed AI to everyone, everywhere, all at once. Here we had a technology where attorneys – not just born-on-the-Internet associates, but senior partners – were asking IT, “What is this?” and “How do we use it?”
Instead of trying to cajole firm leadership into looking at new technology, many technologists and innovation leaders in law firms were in the strange position of trying to get in front of the train, sometimes needing to slow it down to create the appropriate process and evaluation.
We’ve Been Here Before
Novell was founded in 1980. For the next 15 or so years, every year was “The Year of the Network.” Not many partners were wondering in the early 80s when their firm would implement Novell. But it did happen, and one day, the mini-computers had been replaced, all of our PCs were hooked together, and we no longer had to talk about it.
This time, we won’t have to endure a decade and a half of “The Year of GenAI” – it will move much faster. But we’re still at the beginning, and the technology is a lot like what it took to install Novell in those early days: amazing for its time, but also complex, demanding, and expensive when applied on an enterprise level.
GenAI is a lot like networking – or the most extensive network of them all, the Internet – because it’s a core technology. Its real value will come as it is integrated into purpose-built tools designed for specific use cases. When attorneys and staff use email, they don’t think about how it’s a cloud-hosted application that they access via ethernet on their local area network, connected to the Internet, then connected to Microsoft’s server infrastructure, which then traverses in a similar path to the client’s network. It’s just an application that works. We don’t have to “prompt engineer” our way into sending that email. Networking is a built-in enabler of the task at hand.
What’s Next
Microsoft Copilot is a good early example of integrating the capabilities of ChatGPT into existing applications like Office. When it works well, you don’t even have to think about it: Copilot provides additional functionality based on the current context. It’s a rather expensive add-on right now, however, because large language models (LLMs), the core technology behind GenAI, are resource-intensive and expensive to run.
This is a fast-moving space, and that problem is already being tackled. The somewhat ironic “small” LLMs with more targeted use cases hold a lot of promise: a future of more narrow-use, task-specific models that are much less expensive and realistic to train on custom data.
In other words, GenAI designed for legal-specific use cases, tuned by your firm’s work product, built into products or categories you already use instead of a new, standalone application.
What to Do Now
Matthew Prince, CEO of Cloudflare, thinks most companies investing in AI today are just “lighting money on fire.” Like a lot of people talking about AI these days, he means GenAI, and he’s referring to the challenges of putting cutting-edge technology to work in mainstream enterprise use cases today.
Options:
• Ignore this new technology and pretend nothing is happening: this will work for a while, but it’s not a great long-term strategy.
• Keep your finger on the pulse of developments: this is the minimum and a good place for a lot of firms.
• Prepare: There is a lot of valuable groundwork to prepare for a world where your data is more valuable, more accessible, and more easily applied. This is a good long-term investment regardless of how the specifics of GenAI play out.
• Learn and experiment: see how GenAI applies to your firm and practice areas. There are a lot of firms doing this, and it doesn’t require a significant investment. They’ll be well-positioned when new developments come down the road.
• Dive in: invest in GenAI with the goal of creating an early market advantage and leveraging the firm’s expertise to build a moat. This is not for the faint of heart.
Bottomline
I’d advise slow and steady. We’re a long way from the finish line.
It’s wise to experiment with GenAI and be cognizant of new developments, but I wouldn’t feel left out if you don’t have a significant effort underway. For most firms, waiting until there are more mature products with well-defined use cases and identifiable ROI is the reasonable course. If I were responsible for making investments in a law firm today, it would be in preparation, particularly around KM.
#ArtificialIntelligence
#LitigationSupport
#PracticeManagementandPracticeSupport
#FutureandEmergingTechnologies
#Firm