HOW DO WE LEARN TO WORK WITH INTELLIGENT MACHINES?
WHAT'S IT ABOUT?
Overview from TED
The path to skill around the globe has been the same for thousands of years: train under an expert and take on small, easy tasks before progressing to riskier, harder ones. But right now, we're handling AI in a way that blocks that path - and sacrificing learning in our quest for productivity, says organisational ethnographer Matt Beane.
He shares a vision that flips the current story into one of distributed, machine-enhanced mentorship that takes full advantage of AI's capabilities while enhancing our skills at the same time.
"Today’s problems demand we do better to create work that takes full advantage of AI’s amazing capabilities while enhancing our skills as we do it."
Matt Beane is an Assistant Professor in the Technology Management Program at the University of California, Santa Barbara and a Research Affiliate with MIT's Institute for the Digital Economy.
So much time is spent fixated on the “future of AI” that it’s often easy to overlook the present. Global pandemics aside, changes to the way we live and work are typically incremental. So much so, that we often fail to notice the significant shifts occurring around us. It is only when looking back – in a few hundred years’ time, say – that we will be able to recognise the early 21st century for what it is: a pivotal phase of transition.
Consider the boom of new and innovative manufacturing processes in the late 18th and early 19th centuries. We know it as the “Industrial Revolution”. And whilst the term was purportedly coined by French envoy Louis-Guillaume Otto in 1799, that doesn’t mean the ordinary folk of the era held any concept of its lasting impact. In much the same way, us Brits are all too familiar with the portmanteau “Brexit” without truly understanding its implications for future generations.
Only when reviewing entire narrative arcs can we gain a full understanding of scale and impact, and in this case - when it comes to how humans work alongside intelligent machines - we’re still very much at the start.
In this TED Talk, Matt Beane seeks to alter the narrative of the AI Age before it’s too late.
Over the course of several years, a study undertaken by Beane has uncovered a looming crisis that threatens our age-old methods of learning. In his own words:
We’re sacrificing learning in our quest for productivity.
Maybe not tomorrow, or in ten or twenty years’ time – but one day (without intervention) the emerging chasm between old practices and new could sink our society quicker than you can say “Iceberg ahead!” If the surgeons of tomorrow are unable to learn traditional techniques, what hope is there if the machines fail? How can a society thrive – or indeed, survive – with only AI expertise to rely on? At what point do human beings become superfluous?
Beane commenced his study with one big, open question: “How do we learn to work with intelligent machines?”
I’m eager to find out…
SEE ONE, DO ONE, TEACH ONE
Beane opens his talk with an anecdote about “Kristen” – a surgical resident wheeling her prostate patient into the operating theatre at 6.30am. By 8.15am, Kristen is closing the patient, having undertaken a lead role in the “nerve-sparing” part of the process under the supervision of the attending surgeon. Beane states:
Kristen feels great. Patient’s going to be fine, and no doubt she’s a better surgeon than she was at 6:30…
….Kristen’s learning to do her job the way that most of us do: watching an expert for a bit, getting involved in easy, safe parts of the work and progressing to riskier and harder tasks as they guide and decide she’s ready.
As Beane highlights, this kind of learning has been the main path to skill for thousands of years.
Whether you call it apprenticeship, mentorship or on-the-job training, the methods are the same. You watch, you learn, you try, you succeed and – in time – you teach. This cycle is present in all areas of life, and is crucial to our development.
Once upon a time, we didn’t know what the big hand and the little hand on a clock represented (arguably some of us still don’t!), but after learning from others – teachers, parents, siblings – we’re able to 1) tell the time, and 2) teach others how to.
Whilst prostate surgery is slightly more complex, the same process applies. Until, that is, the robots join in…
It’s 6:30am again, and Kristen is wheeling another prostate patient in, but this time to the robotic OR. The attending leads attaching a four-armed, thousand-pound robot to the patient. They both rip off their scrubs, head to control consoles 10 or 15 feet away, and Kristen just watches.
The robot allows the attending to do the whole procedure himself, so he basically does. He knows she needs practice. He wants to give her control. But he also knows she’d be slower and make more mistakes, and his patient comes first.
So, Kristen has no hope of getting anywhere near those nerves during this rotation. She’ll be lucky if she operates more than 15 minutes during a four-hour procedure. And she knows that when she slips up, he’ll tap a touch screen, and she’ll be watching again, feeling like a kid in the corner with a dunce cap.
The example may sound over-the-top, but anyone with experience of handling a new and fancy bit of tech can surely relate.
Remember when you got your first smartphone, or that shiny new laptop? The thought of handing it over to your kids to play with (or butter-fingered Ben from HR, for that matter!), would undoubtedly fill you with dread. Perhaps you had an older sibling who would grab the controller to fight The Boss in video games because you were “doing it all wrong”…
Now imagine instead of a mobile phone or a games console, you were handling a multi-million-dollar piece of robotic equipment… Oh, and somebody could literally DIE from a single mistake? After all, “power-ups” don’t exist in real life. You can’t just reset the game to continue.
One thing I find curious, however, is that Kristen is permitted to partake in the “manual” surgery but not the “robotic” one. Surely there’s more to it than worrying about preserving highly expensive equipment?
I covered 18 of the top US teaching hospitals, and the story was the same. Most residents were in Kristen's shoes. They got to “see one” plenty, but the “do one” was barely available. So, they couldn’t struggle, and they weren’t learning.
AI is effectively blocking learning on the job.
Perhaps attending surgeons hold greater trust for technology than their human counterparts? Maybe it’s a question of culpability. Or do they feel they have greater scope to step in and help in a more traditional environment?
Whatever the reason, gaining experience of new tools and techniques is vital for our ongoing development and survival. And not just in the medical field...
TIME TO ACT
Over the course of his study, Beane connected with a growing group of young researchers investigating the role of AI in start-ups, policing, investment banking and online education, amongst others. After hundreds of hours of observation, interviews and shadowing, the team sought to establish patterns from their data. Beane summarises:
No matter the industry, the work, the AI, the story was the same. Organisations were trying harder and harder to get results from AI, and they were peeling learners away from expert work as they did it.
Start-up managers were outsourcing their customer contact. Cops had to learn to deal with crime forecasts without experts’ support. Junior bankers were getting cut out of complex analysis, and professors had to build online courses without help. And the effect of all of this was the same as in surgery. Learning on the job was getting much harder.
You could argue that certain roles could benefit from an AI takeover. Take customer service chat-bots, for instance. In time, these will surely advance to a stage where they are almost indistinguishable from their human counterparts. They will have mastered complex human mannerisms – such as tone of voice and empathy – and can offer timely support and instantaneous resolutions. All whilst saving businesses money. But we’re a long way off that. Customers value the “human touch”, and that desire for one-to-one dialogue with a real human being is unlikely to disappear any time soon – if ever.
But what of teachers? Lawyers? Firefighters? Humanity is a prerequisite for these roles. How can they balance the inevitable influx of intelligent machines with their own needs to learn and develop?
Beane’s closing example provides hope:
Before robots, if you were a bomb disposal technician, you dealt with an IED by walking up to it. A junior officer was hundreds of feet away, so could only watch and help if you decided it was safe and invited them downrange. Now you sit side-by-side in a bomb-proof truck. You both watched the video feed. They control a distant robot, and you guide the work out loud. Trainees learn better than they did before robots.
We can scale this to surgery, start-ups, policing, investment banking, online education and beyond. The good news is we’ve got new tools to do it. The internet and the cloud mean we don’t always need one expert for every trainee, for them to be physically near each other or even to be in the same organization. And we can build AI to help: to coach learners as they struggle, to coach experts as they coach and to connect those two groups in smart ways.
Perhaps instead of viewing intelligent machines as competition, we see opportunities for collaboration. AI is not designed to “steal our jobs”, but to enhance them; to allow us to achieve things that wouldn’t be possible through the power of humans alone.
Put like that, perhaps the “AI Age” doesn’t sound so scary after all…