This is the third talk from the the ILTA Session – Legal Technology Innovation – Bolstering and Destroying the Legal Profession. This is from Michael Mills, Co-Founder and Chief Strategy Officer at Neota Logic.
I decided to come at the technology question from the human side, to speculate about what humans are still good for in a technology-saturated world of legal services.
I concede. I am obsolete. The robots are winning.
Fastcase is a better legal researcher than I am—despite a University of Chicago law degree, a federal court clerkship, and a hand in hundreds of briefs and memos.
Recommind Axcelerate is a better document reviewer than I am—despite tutelage by demi-gods of the American bar, and years of experience, some of it in unheated warehouses and abandoned salt mines.
And of course Google is a better driver than I am.
Nonetheless, for a while, I have work to do.
Kira is not a better contract analyzer than I am—my pattern-recognizing brain is more precise, more adaptable, and faster than Kira’s algorithms. And the algorithms need training, so I can have a job as an algo trainer—like a dog trainer, but without a whistle or a biscuit.
But … Kira’s algorithms are getting better and its computers are getting faster. My brain is not, alas.
So, one day . . . poof! ZMP for me—that’s Zero Marginal Product, the economists’ term for adding no value at all.
As Harvard professor Bill Bossert said many years ago—“If you’re afraid that you might be replaced by a computer, then you probably can be—and should be.”
Or, as I say to law firm partners who worry that Neota Logic expert systems will cannibalize their billable hour work—“If your business model is to do work that my software can do … you’d better get a new one.”
So what’s left for me? For us? We went to law school, we’re nice people, we’re pretty smart.
Fortunately, there are some things for us humans to do:
Geoff Colvin of Fortune Magazine just published a book with a great title “Humans are Underrated.” (In some contexts, one might argue that he has that backward.)
He writes that the right question is this:
“What are the activities that we humans, driven by our deepest nature or by the realities of daily life, will simply insist be performed by other humans, regardless of what computers can do?”
He then says that the foundation of all the other abilities that make people valuable as technology advances is … empathy.
Discerning what some other person is thinking and feeling, and responding in some appropriate way.
We have evolved to do that—collaboration was essential for survival, in hunter-gatherer and then in agricultural economies.
As Colvin puts the point:
“We want to work with other people in solving problems, tell stories to people and hear stories from them, create new ideas with people. We want to follow human leaders. We want to negotiate important agreements with people, hearing every lilt or lament in their voices, noting when they cross their arms, looking into their eyes.”
We, both individually and as members of groups and organizations, keep changing goals, purposes, understandings, directions, conceptions of the problem, interests—software simply can’t keep up.
People can, groups can.
One might say, then, that what humans will continue to do, so as not to drown in the rolling wave of technology, is what we do best in groups:
- Idea-generation, problem-solving, strategy
- Persuasion, argument, storytelling
So … if groups are essential to our economic survival in a world eaten by software, to use Marc Andreessen’s phrase, how do we know an effective group when we stumble into one?
Cambridge University psychologist Simon Baron-Cohen developed in 1997 a simple test, the RME—Reading the Mind in the Eyes. Participants are asked to choose a word that best describes people’s thoughts or feelings based only on photos of their eyes.
Group members’ average score on RME has proven to be an excellent predictor of group effectiveness.
More recent research supports a much simpler test—no advance testing required. Just count the number of women in the group. More women, more effective. Period.
Another reason for diversity in STEM disciplines!
MIT professor Alex Pentland invented the sociometric badge, a little tag that hangs around your neck and tracks how you work with others—the amount of face-to-face interaction, conversational time, prosodic style, physical proximity to other people, and physical activity levels.
After sociometrically measuring many groups with his little badges, Professor Pentland found that groups do their best work when the participants:
- Generate many ideas in short contributions to conversations. No one natters on.
- Constantly alternate between talking and listening, encouraging, and reacting.
- Take turns.
It does sound a bit like the prescription for a good kindergarten, but it works. These 3 factors are as important to group effectiveness as all others together—individual intelligence, technical skill, personalities, and so on.
Interestingly, and here we technophiles should take note, this research suggests that online, technology-mediated collaboration is far less effective than we think.
Apple agrees—their new headquarters is gigantic, in order to bring people together, physically, to do the empathy thing, to do the human thing.
Google agrees—they engineered the cafeteria (it’s a metrics-driven company) – optimum wait time in line 3–4 minutes, table spacing to encourage bumping, long tables to encourage sitting with people you don’t know.
So, even at Google, there is room for us humans.
But … and here I think we come to the rough reality of the legal services industry (and others too, which raise profound long-term questions about the civic compact)—technology is pushing the performance bar for humans ever higher, chopping off the bottom tail of the bell curve, shrinking the space in which “just OK” is OK, in which being “pretty good” is good enough. It isn’t any more.