AI Revolution for In-House Legal Teams: The Seven Headlines All Lawyers Need to Know

In case you missed it or want to revisit this important topic: Experts break down the implications of Artificial Intelligence (AI) for legal departments in 2023.

Here’s what’s not new news: Read any number of headlines from this year and you’ll know that artificial intelligence is now permeating virtually every industry. 

Here’s what is new news: Contrary to the industry’s typical late adoption tendencies, legal is actually at the forefront of this AI-led disruption. 

The latest installment of Axiom’s “Higher Bar” webinar series examined how generative AI will further impact legal departments in 2023. In discussions of the trends afoot and their relevance to the in-house team, expert panelists including Ken Rubenstein, former head of Governance and Programs and Responsible Innovation at Google, Jason Anderman, current Head of Legal and Compliance at Certa, and Zach Abramowitz, CEO of Killer Whale Strategies and LegalTech expert, revealed seven headline findings:

1. Legal is a “blue ocean” industry.
2.  ChatGPT is an “every” lawyer tool.
3.  ChatGPT is an enabling technology, not a replacing technology.
4.   Lawyers who are not yet using it will be soon.
5.  AI can make legal a better partner to the business.
6.  Lawyers need to embrace the benefits to legal work but beware of the risks.
7.  GCs must think through AI’s legal implications beyond the legal department… now.
   
   

Headline 1: Legal is a “blue ocean” industry.

While usually the last industry to embrace innovation, legal has become a leading AI pioneer. That’s not anecdotal – it’s backed by data. The OpenAI Startup Fund has a mission to invest $100 million in fields where artificial intelligence can have a transformative effect and where AI tools can empower people by helping them be more productive. Tellingly, its initial fund includes four legal adjacent/related companies. Legal may be one of the richest areas for AI disruption, if done right.    

Why? Because “law is a language in/language out business.” Generative AI like ChatGPT, which can digest documents and then produce coherent/original text, can therefore be an industry game changer. And Goldman Sachs says it will be. In a recent report, the firm forecasted that generative AI tools like ChatGPT will accelerate legal industry automation so much so that 44% of all current manual legal tasks may soon be automated. 

Ken has firsthand knowledge of how AI can affect lawyers and legal departments and saw this wave coming years before it was trendy. At Google, back in 2016, he started to plan for issues that were coming over the horizon and he realized the future of AI would one day become very prominent. Shortly after, he helped author the initial AI principles at Google. He thinks the basic principles used then are still pretty consistent today, even with all the changes AI has brought.

Headline 2: ChatGPT is an “every” lawyer tool.

Unlike previous corporate-led LegalTech, because of its ubiquity, accessibility, and virality, ChatGPT is an “every” lawyer technology. “Everyone is reading the news, everyone is seeing these reports, everyone is playing around with this technology—and lawyers are aware that this technology can be immediately and intuitively impactful,” Zach comments.

What are lawyers using it for? “Some of my friends started out doing silly things with it, like drafting an epic rap battle between patents and copyrights,” Ken shares. “And you know, it produces something that's somewhat laughable. But then they started doing more interesting and work-applicable things: case research, clauses, issue lists, and rewriting emails. The more they are using it, the more they are finding it can serve as an enabling tool. In fact, it can be a very good quasi-assistant.

Headline 3: ChatGPT is an enabling technology, not a replacing technology.

Early AI mistakes and pitfalls—such as the lawyers who used ChatGPT to submit case briefs, only to find they contained fabricated case precedents—suggest its limitations when used to replace human production and value. “It’s not a replacement,” Ken adds. “I think that's where some of the mistakes are happening. But if you can use it as a tool, the two of you together can be extremely powerful.”

The right way to look at its enabling capacity right now is to view AI as a starting point. Let’s take writing an expert witness agreement as an example. As Ken had never done this before, he gave ChatGPT a crack at it: “There were two places where it was inconsistent internally and was contradicting itself. But for a lot of it, it was pretty damn good. If I got that first draft from my junior associate, I would not have been upset. And there were some insights in there that I might not have thought of. So, using it as a tool to help me draft? I'm finding it extremely helpful.” 

And there are more starting point use cases. “The other day,” Jason shares, “I was debating with some friends about key ethics for the right use of an LLM for transactional work, and then I thought, wouldn't it be fun to ask ChatGPT to write a policy on the ethics of using ChatGPT for transactional work? And you know what, it was a good starting point. I had lots of edits, but it's a perfect example of what we're talking about here. I don't think it's replacing lawyers anytime soon.” Instead, it is helping lawyers push forward their starting point further down the line.

“What I'm hoping it will do is create more white space in our day to tackle high-risk, high-value matters while it handles more of the low-risk, low-value issues."

The question legal will have to tackle is: how far down the line can lawyers actually push that starting point?

Headline 4: Lawyers who are not yet using it will be soon.

The record-breaking rate of adoption means that this tool will have a ubiquity that will make it impossible to ignore or escape from. The rate of adoption and progress here is going to be something that we're simply not used to as humans.

Zach likens the surge to social media, but more accelerated and more historic: “ChatGPT wasn't just the quickest product to get to one million users, which took only four days. It was also the quickest to get to 100 million users, which it did in two months.” 

What this means is that even tech-skeptics should learn enough to ensure they are not being left behind. What about those lawyers who cite the early mistakes and pitfalls?  

They need to understand that the rate of progress and improvement is simply astonishing. In fact, engineers are taking those early pitfalls, learning from lawyers, and informing tool tweaks which are progressing tool sophistication faster than almost any category of technology prior. 

Jason adds, “Legal departments will see that over the next couple of years, the disruption of AI will not move at a linear rate; it will be exponential. Lawyers should begin experimenting now: learn how to be a prompt engineer yourself, and how to ask the right questions to get the right answers out of these tools, because in a couple of years, it's going to be fundamentally part of your workflow."

Headline 5: AI can make legal a better partner to the business.

When legal increases its speed, it can enable accelerated revenue capture. That’s how legal becomes a better partner to the business—by increasing its cadence without sacrificing its competency. AI holds the promise to help legal do just that. 

Jason recommends you think long and hard about the most reasonable approach to provide a measurable impact. He suggests lawyers should consider drafting contracts via business intelligence roles with a Q&A. It will tell you, for instance, every issue you would need to spot on a limitation liability clause and give you the backup of the backup of the backup to close the deal in a way that’s already approved by everyone. By using AI this way, you don't have to escalate and get permissions when you're negotiating and can amp up your ability to close based on your bargaining power, while customizing your drafting to that particular deal at hand. As opposed to what they typically do, which is to use these incredibly aggressive Microsoft Word templates with a million versions back and forth in a slow sales or procurement cycle time. 

He says, “Instead, what if we could use generative AI to search all of our templates, playbooks, escalation procedures, and thousands of signed contracts in our repository to automatically build that Q&A? Then we just have to edit it to get it in place. And we can constantly re-edit in the future based on new, cool things we learned in negotiating that close deals faster. We now have way lower risk management, and our contracts will be drafted much better.”

Headline 6: Lawyers need to embrace the benefits to legal work but beware of the risks.

We don’t need to tell that to lawyers, right? A lawyer’s number one job is risk mitigation, so we know you’re always looking for the risks and weighing them against the rewards. And make no mistake, as promising as the latter category is, consider also how potentially damaging it can also be.

But you can’t just ban it. Social media comes with enormous risks, but companies have learned to embrace and mitigate those risks as they’ve become woven into the fabric of society. The same will be true of AI. In-house lawyers must therefore be mindful of its risks in order to mitigate them, as they can never be eliminated altogether.

Three important risk-mitigating tips when using AI:

1. Don't humanize chatbots and generative AI tools. They are not intelligent, although they may have the appearance of intelligence.
2. Limit its use to being a tool or assistant. Don’t view it as a replacement: Remember these tools currently “hallucinate”/speak convincingly on topics that are just not true.
3. Evaluate all potential risks for your department. The AI tools will become much more valuable as they are trained on existing legal department precedents (contracts, pleadings, memos, etc.). Legal departments will need to work through issues of privacy/client privilege in providing confidential and identifiable information to these tools.

Headline 7: GCs must think through AI’s legal implications beyond the legal department… now.

And for that, they will need help. AI is, after all, a novel area of the law. Axiom has been providing lawyers who specialize in AI-related issues for several years now. We recognize that the development of AI presents both incredible opportunities and significant challenges for lawyers. We’ve covered the former above. The latter is where Axiom can really add value. As algorithms make decisions that directly impact people, lawyers are facing new questions surrounding the ethical use of AI technology and legal compliance. And Axiom has legal talent well-versed and practiced in this emerging area of the law. 

Here’s how Axiom can help:

• Axiom provides flexible lawyers with experience consulting on issues related to social responsibility, including ethics, fairness, and removing bias in algorithms.
• Another key challenge in-house teams face when implementing AI is regulatory compliance. Axiom’s lawyers can help companies navigate relevant regulations, including but not limited to, the EU AI Act. 
• Data privacy & cybersecurity also come into play, and Axiom has the talent to support the legal department as it assesses and addresses privacy policies and procedures.
• Finally, the implementation of AI raises questions about intellectual property. Our attorneys have experience consulting on issues related to determining ownership of AI-generated works and copywriting code used to train AI models.

💡 Learn how to better manage the unique challenges and complexities presented when onboarding, utilizing, or developing artificial intelligence.