Matt Beane: How do we learn to work with intelligent machines?
马特·比恩: 我们如何学习使用智能机器?
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. Full bio
Double-click the English transcript below to play the video.
her prostate patient into the OR.
她的前列腺病人进手术室。
some of the nerve-sparing,
参与进行神经保留手术,
以让病人恢复勃起的功能。
that can preserve erectile function.
though, but he's not there yet.
但那会儿他并不在手术室。
给病人打了麻醉。
切开了一道8英寸的切口,
incision in the lower abdomen.
she tells the nurse to call the attending.
便让护士打电话给主治医生。
are mostly in that patient --
他们四只手都在病人体内,
but Kristin leading the way.
克里斯汀则主导了手术。
he let Kristen do a little nerve sparing),
让她进行了部分神经保留的操作,
初级住院医生的协助下
looking over her shoulder.
给病人做了最后的缝合。
the final line of sutures.
自己更进了一步。
than she was at 6:30.
the way that most of us do:
其实和我们的并无分别,
safe parts of the work
and harder tasks
并且有专家在一旁指导。
by this kind of learning.
part of what makes us human.
coaching, mentorship, on the job training.
学艺、训练、教导和在职培训,
“see one, do one, teach one.”
这被称为“看、做、教”,
在培养人才时所用的方式。
around the globe for thousands of years.
in a way that blocks that path.
方法却反其道而行之。
in our quest for productivity.
牺牲了学习必经的过程。
while I was at MIT,
第一次发现了这个现象,
it’s happening all over,
以及各项人工智能的应用场景中。
and with very different kinds of AI.
are going to hit a brick wall
在学习如何掌握人工智能时,
另一个前列腺病人进手术室。
is wheeling another prostate patient in,
机器人连接到病人身上,
robot to the patient.
10 or 15 feet away,
to do the whole procedure himself,
主治医生独自便可完成手术,
and make more mistakes,
操作得更慢,还有失误的风险,
near those nerves during this rotation.
完全没有机会碰到病人的神经,
15 minutes during a four-hour procedure.
操刀超过一刻钟就算是走运了,
and she’ll be watching again,
她又不得不回到观察者的角色,
with a dunce cap.
I’ve done in the last eight years,
所有关于机器人的研究一样,
我也提出了一个宏大的问题:
with a big, open question:
with intelligent machines?
observing dozens of residents and surgeons
观察了数位外科医生和住院医生。
interviewing them
也做自动化手术,
with the residents as they tried to learn.
US teaching hospitals,
美国18所顶级的教学医院,
and they weren’t learning.
I needed to know how widespread it was:
但我想知道,这样的现象有多普遍?
blocking learning on the job?
人工智能阻碍了人们的学习?
but growing group of young researchers
年轻但正迅速成长的研究团队。
of work involving AI
关于人工智能的实地研究,
like start-ups, policing,
用了数百个小时进行观察
and many hundreds of hours observing,
side-by-side with the people they studied.
和他们一起生活、工作。
我想从中寻找出规律。
the AI, the story was the same.
人工智能,结果都非常相似。
and harder to get results from AI,
想从人工智能中获益,
从专业工作中脱离出来。
expert work as they did it.
their customer contact.
联系消费者的工作外包出去,
forecasts without experts support.
去做犯罪预测工作,
cut out of complex analysis,
online courses without help.
was the same as in surgery.
和上述外科例子是一样的,
was getting much harder.
a billion and a billion of us
我们中有5亿到10亿人,
适应人工智能。
in our daily work by 2030.
that on-the-job learning
在职学习机制将一直存在,
that most workers learned key skills
多数员工在工作时才能真正掌握技能,
potential future impact,
人工智能对未来潜在的影响,
that may matter most right now
that blocks learning on the job
我们目前最需要的东西。
a small minority found a way to learn.
找到了学习的方法,
so they bent and broke rules
所以他们要改变和突破规则,
在医学院时可以参与到自动化手术中,
in robotic surgery in medical school
of their generalist education.
with simulators and recordings of surgery,
研究模拟器和手术记录,
they found ways to struggle
他们找到了奋斗的方法,
with limited expert supervision.
because it bends the rules
因为它修改了规则,
because it gets results.
因为这样的学习的确有效。
the star pupils of the bunch.
and it’s not sustainable.
也并不可持续,
they need to do their job.
struggle and challenge in their work
遇到的困难和挑战,
to tackle hard problems
there was an expert nearby
足够的专家资源指导他们,
against catastrophe.
of struggle and expert support
I could get of this on the ground.
你经常要直接处理简单易爆装置,
you dealt with an IED by walking up to it.
hundreds of feet away,
你觉得安全的时候才能提供帮助,
if you decided it was safe
in a bomb-proof truck.
而你大声地指挥工作,
and you guide the work out loud.
有更好的机会学习。
did before robots.
start-ups, policing,
外科手术、初创企业、治安系统、
online education and beyond.
we’ve got new tools to do it.
更好的工具辅助学习,
always need one expert for every trainee,
需要专家进行一对一、面对面的教学,
不需要在同一个机构中。
or even to be in the same organization.
to coach experts as they coach
还可以指导专家进行更有效的教学,
in smart ways.
on systems like this,
有这样的教学系统,
on formal training.
关注入职培训,
is in on-the-job learning.
在职培训当中。
要求我们要做得更好,
of AI’s amazing capabilities
人工智能带来的便利,
I dreamed of as a kid.
这一未来的最佳时机。
ABOUT THE SPEAKER
Matt Beane - Organizational ethnographerMatt 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.
Why you should listen
Matt Beane does field research on work involving robots to help us understand the implications of intelligent machines for the broader world of work. Any of his projects mean many hundreds of hours -- sometimes years -- watching, interviewing and often working side by side with people trying to work with robots to get their jobs done.
Beane has studied robotic surgery, robotic materials transport and robotic telepresence in healthcare, elder care and knowledge work. He has published in top management journals such as Administrative Science Quarterly, he was selected in 2012 as a Human Robot Interaction Pioneer and is a regular contributor to popular outlets such as Wired, MIT Technology Review, TechCrunch, Forbes and Robohub. He also took a two-year hiatus from his doctoral studies to help found and fund Humatics, an MIT-connected, full-stack IoT startup.
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. He received his PhD from the MIT Sloan School of Management.
Matt Beane | Speaker | TED.com