How Might AI Impact the Future of Legal Practice?

A Twitter survey earlier this spring asked, “Are you concerned about Artificial Intelligence?”  My reply:

One of the over-arching themes at Social Media Marketing World centered on AI – Artificial Intelligence.  No surprise, AI has overtaken most conversations regarding politics, The Future, science and technology, and even the future practice of law.

Here, I summarize notes from the Thursday morning session led by Christopher S. Penn @CSPenn, and the Friday morning Keynote panel, of which Penn was a member, along with Sandy Carter of Ecosystems, Christopher Carfi of GoDaddy, and Bryan Kramer of Pure Matter as moderator.  If you’re in a rush, you can skip to the Conclusion for Lawyers at the bottom.

Marketing is the over-arching, big umbrella under which various activities fall, including:  sales, social media, SEO, market research, customer insights, media buying, and more.  For decades, the Brand has pushed content to Consumers, yet with social media, Consumers are in the driver’s seat, and savvy Brands now employ social media listening tools to follow the interests of their audiences and serve back to them what they tell the Brands they want.  Fast-forward

Cognitive Marketing
Penn opened his presentation with a quick state of the nation:  marketers lack agility.  Marketers and marketing tools don’t learn fast enough (anymore).  Solution?  Cognitive Marketing – marketing that learns using AI technologies.  Penn explained, “AI tech is a lot like humans – only it evolves faster and better.  Fundamentally, an algorithm is a set of rules that are repeatable and performed by a computer.  All social media is an algorithm.”  An algorithm is a bit like a recipe – you oversee the ingredients based on what you want for the outcome.  The magic happens with Machine Learning.

Penn illustrated his human learning/machine learning parallel:  “Hand a toddler a pile of blocks, and he will start to learn – trying to figure out how the blocks fit together, what patterns can be created.  That’s independent learning.  If an adult stands over the toddler holding a red block and repeats, “This is RED,” over and over – the child will begin to learn to associate the word and the color.  This is supervised learning.  Supervised learning teaches a machine to recognize patterns.  We teach a machine, ‘This is my Twitter handle,’ and it begins to recognize that information within other text.  Machines need help, though, recognizing good sentiment versus bad sentiment.”  (Here, Penn refers to a simple machine instruction to look for your company name.  It will turn up that name, regardless of whether the surrounding context is positive or negative.)

Examples of Machine Learning
Penn explained that Machine Learning consists mostly of math and statistics, and Deep Learning refers to many layers of machine learning.  For example:  

A woman fighting cancer ran into difficulties when her physicians tried several modes of treatment without optimal impact.  Her genome was fed into IBM Watson’s Oncology Expert Advisor, which it cross-referenced against 220,000 digests on chemotherapy.  Based on Watson’s recommendation, her treatment was changed and she went into remission.  Watson’s effort?  Took 11 minutes.

If you use Google Translate, you are familiar with the process of entering one language and Google translates to another, such as English to German.  A few years ago, the result was a rough approximation of correct translation – some words and phrases were inaccurate.  Yet, a few months ago, the results from Google Translate were noticeably more on-target.  What happened?  Machine learning.  Google ingested all human languages, developed its own Google language, and now translations go through an additional step before rendering results:  English to Google to German.  Penn commented, “Google is reverse-engineering the Tower of Babel.”

Adding Emotion to the Mix
Penn noted that we can present a large set of data and ask Google, “What do these things have in common?”  The system will generate the common thread.  Now, we can write an article about that theme.  His example:  In the last year, if you have followed Fantasy Football or Minor League Baseball, you have not read one article generated by a human.  Think about it – much of sports is driven by statistics, fodder for a machine to spit out data-driven content.  But, we have to be careful when the bias of the person entering the data colors the final product.  Sandy Carter touched on this with the old adage – “When it comes to data, 'garbage in' often means 'garbage out.'  Businesses must prepare for AI by collecting the best data.”

Because we are at the stage of technological development where machines can handle Natural Language (old search: “restaurant BBQ Dallas”; search now: “What's the name of the barbeque restaurant near UT Southwestern in Dallas”) – we can understand more of the context surrounding the query and thus “fingerprint the emotion of the author.”  This development enables machines to use image recognition and find new commonalities, it also gives rise to the proliferation of Chat Bots that provide customized solutions for customers.

Do You Want Quinoa With That?
Anything that’s rule-based can be automated. “If this, then that,” is at the root of rule-writing for code (the 50-year old BASIC programming language was rooted in IFFT for quick usability), and it’s also fundamental to decisions lawyers make every day.  During AI discussion from the Friday panel, Carter gave an example of fully-automated restaurants like Eatsa, where no humans are involved in the food-service side of the transaction.

She also spoke of a store in Japan where the store’s (automation) can measure a shopper’s intention when the shopper enters the store – are they browsing, in a rush, following a routine, etc.  Imagine how that might help you assess a prospect.

Carter uses AI now to inform her advertising buys, and to leverage her time.  A bot may provide her with suggestions for optimal buys, but she makes the final decision.  This is an example of Christopher Carfi’s belief that a portion of the population will run the machines (and others will be run by machines).  Carter suggested several tools, including GrowthBot – useful for investigating the keywords companies are buying.  

Conclusion for Lawyers

The legal landscape will continue to evolve, and rapidly, because of a number of factors – AI is merely one.  A few significant (points) are:

1.    Automation and machine learning are already here.  How can you learn to think like a machine, or think in an ordered fashion regarding the data you need, the data your clients need, and the data you both produce?  Identify services that will enable you to automate the gathering and synthesis of information, especially with methods that are repeatable – as those processes can be templated, taught to others -- increasing efficiency.

As attendees from DigimindCI captured:  "The biggest takeaway (from Social Media Marketing World)…not only can AI make life easier for consumers, which is always the end goal, embracing AI can also make life easier for marketers, which facilitates the end goal."  

Whether you like the term “marketer” or not – you are effectively in that role because you are the ultimate driver to promote your business.  To do so, you must learn from your audience (clients and prospect base).  AI allows this and is scalable.  So, while the Big Firms take advantage of ROSS, there will be new market entrants that assist medium-sized and small firms as well as solos.  Plus:  you could create your own solutions, if you are so inclined.  Isn’t that exciting??

2.    Start NOW to uncover ways that you can think and create to make use of the machines and tools coming your way.  Build your skill set and that of your team so that you broaden your capabilities and thereby avoid silos and obsolescence.  Start NOW locating the tools necessary to increase your knowledge.  I asked Penn how Marketers with strengths in writing and content might shore up their abilities in math and stats.  He suggested studying R or Python, two programming languages for data science.  If you feel a bit fish-out-of-water regarding learning code, check out BASIC.  As noted above, the simple logic is akin to the cognitive steps employed by lawyers every day.    Set your end-goal not to be a coder, but an “Algorithmic Thinker” because, as Penn stated, this will require you to think like coders, designers, and architects – designing systems in your head.

Microsoft Cognitive lets people build Apps fairly easily on a number of measurable criteria:

3.    (2a.) Penn showed a list of current job roles  that will likely be overtaken by machines in the future.  

He pointed out, however, that if you were able to deliver in more than one area – you would be less susceptible to replacement than if you had a narrow tool set.  
Pair this with the concept that those who deepen their niche of knowledge -- Subject Matter Expertise -- will drive more demand generation, stronger lead funnels, and ultimately gobs of data to process for knowledge around audience and deliverables, thus strengthening their future prospects far beyond those who are generalist practitioners with only one or two of the skills noted above.

4.    Leverage social NOW.  If you haven’t yet fully utilized social, or haven’t even bothered with social, ask for help.  Find an ally to teach you the basic platforms (Linkedin, Twitter, Facebook) and how to publish your own story.  Then, teach yourself or seek tutoring/training (some can be found online via YouTube videos, for example) – so that you understand how analytics work, what to look for, what to measure, and what to do with the results of those measurements.  The results will inform your next moves – thus making enhanced use of your time and your budget – as well as offering information that amplifies your unique message.

5.    Leverage AI for greater efficiency.  Lawyers are always pressed for time and for ways to gain time-efficiency.  For the portion of your practice where you repeat processes and repeat answers to common questions – you can build bots and automated services to facilitate responses while gaining efficiency.  Anything for which you have a template can be automated.

6.    Penn proposed that the future will include only three primary job functions:  Developers (who build the bots, programs, and write the code in order to reach goals), Data Scientists (people who can create and process the data and supervise machines on good or bad output), and Marketing Technologists (those who can convert data into Business Results).  Which one are you?  Which one will you be?  

7.    Here’s is Penn’s summary of the SMMW event.    His decision tree interests me because he observed that Social Media Marketers are missing the scope of what’s coming.  

This is important to lawyers because:  very few legal marketers have sufficient awareness of what’s coming, let alone sufficient understanding of its application to legal practice and the business of law.
— (My two cents.)

Add to this, Gabriel Teninbaum cited a recent Altman survey:  "Nearly all partners surveyed think tech is here to stay...yet only 4% evaluate their partnership as highly prepared to deal with the new reality."

Penn summarizes SMMW in another article – and also emphasizes the high value of storytelling that is strong enough to withstand numerous social platforms.   I believe this is a crucial point for attorneys because the social age has opened the doors for individuals to tell their own stories in their own ways.  Legal professionals who are adept at crafting content with (the right amounts) of brevity and creativity ingredients will become the next "Top Chef" of their domain in the legal sphere.

A quick way to keep up is to follow thought leaders on Twitter.  Be a fly-on-the-wall to their discussions as they hash out the pros and cons of a variety of topics that will impact law now and in the near future -- even join in with your ideas!  I keep a list of people whose tweets and articles relate to "Future Law."  A few Twitter handles:  @CSPenn (of course); Dan Lear, Jordan Furlong, Ed Walters, Ron Friedman, Gabriel Teninbaum, Heidi Gardner, Mark Cohen, Daniel Martin Katz, Frank Paquale.  There are many more, but these are a good start.

I’ll be at Stanford this week for the @CodeX Future Law event.  If this English major can decipher what they discuss, I’ll tweet & write about it.

What Do You Want to Know Next?  Send a note with a topic suggestion.

2017 SPRING LEGAL MINDSTORM – Come early for the Ethics CLE, stay for the interactive discussion of AI, Blockchain, Collaboration, Millennials, Early Exits and other topics shaping and inspiring the future of legal practice – and legal innovations.  REGISTER before April 25th for Early Bird pricing.

Flip Cat Consulting works with law firms, practice groups, individual attorneys, and other professionals to design marketing and business development strategy.  We work onsite or remotely, from specific projects to global change management.  Contact us to arrange a free consultation.