Pratique du Shadowing: How AI Is Changing Strategy—and 4 Skills Leaders Need Next - Apprendre l'anglais à l'oral avec la vidéo
Choisissez un mode d'entraînement
Chaque mode ouvre une salle d'entraînement en plein écran.
553 phrases
Si les phrases sont trop courtes ou trop longues, cliquez sur Edit pour les ajuster.
1
One of the most depressing things I see
2
when I talk to boards of very large companies
3
and to C -level leaders is a lot of them have come to believe they just can't change, right?
4
So I show them what AI can do and they're like, well, I guess we'll ride this out.
5
And I'm like, that's an insane answer.
6
You have 100 ,000 people working for you.
7
The smartest people in the world are working for your company.
8
You have customers and resources.
9
The idea that you can't be flexible compete seems to be overly baked in right now.
10
Ethan Mollick, professor at Wharton and author of Coexistence, welcome to the IdeaCast.
11
Thanks for having me.
12
Yeah.
13
So you've been closely tracking advances in AI and constantly experimenting with all the new tools.
14
What is different today than even a few months ago in terms of how new developments should guide the way
15
that leaders are thinking about strategy in particular?
16
One is the exponential curve continues.
17
AI just keeps getting better.
18
There isn't really a sign of that stopping right now.
19
And the second is that agents are real,
20
which are AI tools that can go off and do work on their own autonomously and do real process work, real important things. And
21
that has become a viable approach to work in the near past
22
it seems like a lot of companies right
23
now are thinking of ai as really a productivity exercise you know you're automating tasks you're saving time you're reducing costs
24
how does you know a genuinely strategic ai transformation strategy differ from
25
that what do you think companies should be doing now i think
26
it's actually pretty urgent to do a few things one is viewing
27
this purely as a productivity tool gets you in trouble in a lot of different ways.
28
People are getting ROI from AI, right?
29
They're getting ROI from that productivity gain.
30
Productivity gains are real.
31
The problem is, is that those productivity gains get eaten up by the organization, which is if you don't change any part of the process,
32
AI is mostly a single player entity and you end up with more of everything.
33
So if you don't change your KPIs, people can produce more PowerPoint, they produce more code.
34
That's not actually what you want.
35
So leadership needs to decide what this thing is.
36
And the second is AI is getting very capable.
37
And I worry that if our default is automation, we end up with a situation where we just keep handing tasks to AI
38
and don't think hard enough about where humans fall into this process.
39
And that leads to a lot of danger as well.
40
Is your argument
41
that leaders need to move from asking sort of where can we use AI in this existing business alongside humans
42
to a question that's more like, you know, how does AI change our business going forward?
43
At this point, we have ample evidence
44
that AI can regularly do the kind of work
45
that would take a human two to 18 weeks in a reasonable amount of time, a reasonable amount of money with high success rates.
46
That implies a lot of different things.
47
It implies that what your organization does has to change.
48
How you think about the boundaries between jobs is changing.
49
I talk to C -level leaders all the time.
50
And one of the first effects they see of AI, and we saw this also in our research, is that jobs start to blur.
51
Coders do design work.
52
Designers do coding.
53
Product managers do everything.
54
How do you start thinking about what the boundaries of your organization are?
55
What's the product you're trying to produce?
56
Is it an interim product of a PowerPoint?
57
What was the point of the PowerPoint in the first place?
58
You start to run into really fundamental questions about what the goal of the organization is,
59
how you accomplish that, what are your values for getting there?
60
And they're urgent.
61
This isn't sort of theoretical, you know, let's sit down in an academic way.
62
It is your organization is already changing from the inside.
63
What do you do about that?
64
So what are some best practices
65
that you recommend to leaders who are grappling with these issues right now
66
that need to start having those really urgent conversations and not just have them,
67
but sort of make them progress enough so that they can actually start making some real decisions?
68
There's some individual feedback and there's some organizational feedback.
69
On the organization side, the model I've seen works best is sort of like a meta model, which is leadership lab and crowd.
70
So you need three things to succeed.
71
You need the leadership team to actually lead in this, right?
72
That's not just say, use AI.
73
They have to make decisions about how AI is used, what the organization does with AI, how to incentivize people.
74
The crowd means giving people access to these tools we're at a point now we're giving organizations inferior ai access
75
because either just
76
because you haven't bothered to improve the most recent models leads
77
you to trouble like you'd actually become very deluded about what ai can
78
and can't do
79
and then you need a lab
80
you need people doing 24 7 work who are not primarily technical by the way there'll be technical people there
81
but also organizational people who are thinking about how do we restructure the organization around ai
82
they're benchmarking what systems could do building systems of the future, taking ideas from the crowd and ideas from leadership and helping build them.
83
So you need to be doing this as a whole organization response.
84
At the individual level, the factor that I'm seeing that differentiates leadership that succeeds at AI use
85
and organizations that succeed in those that don't is leaders who are willing to eat risk, who are willing to say,
86
look, nobody knows what's going to happen here, but unless we do active experimentation, we're not going to know anything.
87
And if you just let everybody in the organization have vetoes over AI use because it might be risky,
88
it is, but not doing anything is risky too.
89
And that also means actively using it yourself.
90
So leaders who don't use the latest AI systems, who wait for reports from other people, who are waiting for training some week,
91
fall behind those who just are experimenting.
92
So I think the last time that you and I talked about this,
93
you sort of had this model for human -AI collaboration that was either a centaur, you're half using your human skills,
94
and then half using AI or a cyborg, where you're going back and forth between the two.
95
Talk about how that is evolving to one that you now describe as conjurer work,
96
that you think leaders should not only be doing themselves as individuals, but also enabling in their organizations as part of their strategy.
97
The biggest change, if you really look at the sweep of the last year, was that in every independent study we have,
98
basically, AI went from regularly being able to do a half hour of work, 45 minutes of work,
99
to you can get 16 plus hours of work out of a single AI instruction with good chance of success.
100
And that means you don't work next to the AI anymore.
101
You don't work back and forth with it in the same way you did before.
102
You are actually assigning the AI to tasks almost like your team you're assigning and it does the work.
103
So a conjurer is really about conjuring a team into existence when you need it.
104
It's almost like writing an RFP for a project you need done
105
or SOP or any of the other standard TLA three -letter acronyms that we use in business to assign work.
106
And you send that out and the AI just does that work.
107
That's a very different situation than I prompt the AI, look at the results, I prompt it again.
108
This is a very different, you know, action -oriented field.
109
How does that change how leaders approach, you know, the structure of the organization, what it even means to be a manager?
110
So everyone's a manager to some degree right now.
111
Everyone has AI to manage.
112
And AI skills look like management skills.
113
In fact, there's a great study out of Anthropic that shows
114
that the people who are the most successful coders at the first pass of coding are actually managers, not coders
115
because they do a good job specifying what they want and the AI executes on that.
116
So that is the big problem, right?
117
Your organization is something capacity increases.
118
And I think organizations that don't have imagination are going to retreat to automating away the work.
119
And automation is a well -worn playbook in the business world.
120
I think it's a recipe for disaster.
121
I mean, it's a disaster for all of us.
122
If all of our work becomes automated or we just become verifiers of AI content, that's like not a good situation to be in.
123
But also it removes all the differentiation.
124
Like leaders need to be building companies that blend AI
125
and human work together with process and with culture in ways that accomplish things that neither could do alone.
126
If you just move to automation, you're going to have the same Claude -based company everyone else does.
127
Well, that's an interesting point.
128
So, you know, when we're talking about strategy, we're also talking about finding competitive advantage.
129
So if every company has access to the same frontier models,
130
where exactly can leaders differentiate their organizations from competitors?
131
that's that's where the key kind of comes in you are
132
probably not using ai enough inside your organization not trusting enough
133
but you then want to bring people back in i mean
134
we're already seeing as the the most extreme version of this
135
right in terms of organizational change are things like software dark factory
136
so dark factory it's factory all the work is done by robots
137
you don't need humans in it you don't need lights um software dark factories i know one company strong dm
138
that has been shipping product to customers and uh there's no human involved in
139
that their rules are no human can look at code, no human can write code.
140
It is 100 % AI automated.
141
I think that's a dangerous approach because you're just letting the code do the work.
142
I think we need to think about building twilight factories,
143
situations where the AI proactively reaches out to people and brings them in because they have variants and views, because their process or expertise matters,
144
because their sign -off approvals need it, or because they find the work interesting.
145
And I think we need to start restructuring work around the idea
146
that these systems really can accomplish many things and will be able to accomplish more in the near future.
147
So it's integrating it with humans at the right places at the right time
148
and process that's going to give you a differentiator.
149
Where are the specific ways in which you think that as fast as AI is evolving,
150
humans still have a role and particularly leaders still have a role?
151
One thing is you can't really be in the loop in the same way you were, right?
152
It was easy to be in the loop when it was a chatbot.
153
It's not easy when, I mean, I literally have projects where I have a swarm of 100 agents solving a problem for me.
154
Like I have the report out to me like I'm a manager.
155
So I am in the loop to the extent that,
156
you know, they give me a PowerPoint about their progress and they show me demos of what they're building, but that's very different than,
157
you know, I'm approving every action.
158
Right.
159
And so leaders actually, one of the outputs of AI is leaders are super empowered.
160
I keep seeing organizations where leadership basically takes back control over their company because they can now directly ask the AI for,
161
you know, for the information they need.
162
I have talked to people who are like C -level leaders who've actually diagnosed problems of plants
163
that no one brought up to them
164
because they could just literally query all of the information at their company's disposal
165
and ask questions and have the AI go off and do tasks for them.
166
So it's a really interesting change in the span of control if you set this up right,
167
which lets leadership actually.. think more about things rather than hearing things through three or four different layers of leadership.
168
So would you say then that AI agents are making strategic decisions on behalf of leaders and organizations with the right context,
169
you know, given to them?
170
I mean, they already are.
171
If you have an AI agent that is doing any long running work, right?
172
Epoch, which is a AI benchmarking company, was reporting that they could get regularly sort of between two
173
and 18 weeks of engineer's work out of a single prompt to Claude for a couple hundred dollars, right?
174
That involves tons of strategic decisions and judgment calls and creativity calls.
175
They may not be your judgment or creativity, but we can't say anymore the AI isn't making strategic decisions.
176
It is.
177
The question for you is, where do you want to make decisions and where not?
178
And just like any team, you can give it those instructions.
179
You could say, listen, I want you to come with the five best options, your defenses for each of them,
180
surface those, pressure test them for me, and then deliver so I can pick between them.
181
And this is the format I want them in.
182
Or you could say, just go ahead and execute, right?
183
You're hooked up to code, build a prototype for it and test it with a few people.
184
Those are now options you have available to you.
185
So some of this is reconstructing what we've already done with humans.
186
So what's the process by using humans and AI together that they get a working idea that's much better than ideas before?
187
How does that get done much faster and better?
188
Where exactly though, does a leader know where they add the most value and where they don't?
189
Like, what are the points at which they should be involved absolutely every time?
190
I think part of this is you have to make decisions
191
now about how to best accomplish what your company wants to accomplish.
192
And you obviously don't want to give this all up, right?
193
You don't want to give it up because the AI is not as smart as you at what you're best at.
194
Because the AI doesn't have enough variation.
195
Because it doesn't actually absorb your culture in a deep way.
196
You know, you can give some of that as context, but that's part of process design.
197
I mean, we do this all the time.
198
What are the limits of your product engineering group versus your marketing group versus your legal team?
199
We have to think about the same kind of process design here, which is where are those limits and limitations?
200
I think the danger is that by skipping doing that, you either move to an all -automation world where all those choices are made by AI,
201
or you underestimate what these systems can actually do.
202
So it is actually disciplined decision -making that gets you there.
203
And what are the things, the aspects of strategic leadership that you don't think can ever be delegated or outsourced?
204
So I'm always careful about bright lines where we say something can be never done, right?
205
Never, ever.
206
But the systems are jagged.
207
They're good at a lot of stuff and bad at stuff you wouldn't expect.
208
And that means, by the way, there's different things the jobs do.
209
Think about coding, right?
210
You are hiring coders to be the best at writing elegant lines of code.
211
Then AI could write the code.
212
So then it became about sort of systems engineering could you see the bigger picture
213
now ai is increasingly doing some of the systems engineering work in some ways it's very upsetting
214
if you're a coder because the nature of the job is changing in ways
215
that are good or bad all the time right
216
and what you learned is crap no longer is i think
217
we're going to see the same thing about leadership like right
218
now your maybe your judgment calls are about uh
219
you know about the details of pricing strategy the ai might be better pricing video
220
you won't know until your lab tests it for you i can't give you the answer in your industry
221
or not you only know by experimentation and you don't have to do
222
that all alone right that's where your lab and your crowd comes in.
223
Your crowd is surfacing use cases and values of AI.
224
You don't have to do this all yourself.
225
But I think a big danger of companies is they think someone's solved this problem for them.
226
Someone's figured this out.
227
We're all faking it.
228
Nobody knows anything.
229
Like the AI labs don't know anything.
230
The vendors are all making this up.
231
This is a three -year -old technology.
232
So you have to sort of seize control to some extent
233
of this yourself to figure out what you want to use it for.
234
Are there capabilities that you suggest your students, your MBA students, your kids,
235
aspiring leaders that you suggest they develop because you believe they'll be increasingly important in an AI -dominated world?
236
So I think there are four things that are very useful.
237
They're not the only things.
238
One of those is deep knowledge, which is the ability to actually understand a field really well, right?
239
So we're on this podcast, and you know for the beginning of my question or not, before I even say like a few sentences,
240
whether this is going to be an interesting answer or not, I hope this is one of them,
241
and whether you need to redirect things or not.
242
That comes from deep expertise, right?
243
It's hard to make that judgment call if you don't have the expertise.
244
Similarly, in your industry area, you can look at a spreadsheet and know whether something is wrong.
245
You don't have to read those 400 pages of AI content because you know right away because you're an expert.
246
It's one of the reasons why junior people often struggling now.
247
They don't have that expertise.
248
Second thing you need is wide knowledge.
249
As people start to cross field boundaries,
250
knowing a little bit about many things helps you give the AI better feedback and ask for the things you want.
251
The third thing you want is taste you want to be able to like pick among things and
252
that could be company taste too what strategy feels like the
253
kind of strategy our company would do are these many designs the ai come up with
254
which is the right one selecting among many options
255
and i think we develop taste and then the fourth one is agency the willingness to take action under uncertainty.
256
And I think that's a little bit harder to teach.
257
We try and do that.
258
I'm an entrepreneurship professor.
259
But I think giving people the leeway to engage in experimentation is important.
260
How are you encouraging the sort of old guard of leaders
261
who aren't accustomed to such uncertainty to become more comfortable with it?
262
I think there's a danger of people overstating the generational aspect of this technology.
263
Like it certainly was true that digital natives were a thing.
264
Like you didn't grow up with the internet.
265
There's a lot of weird stuff about the internet you don't understand as an internet user.
266
It doesn't work the same way with AI because AI doesn't have the same sort of innate background.
267
It's sort of a single player setup at this point.
268
And because it rewards expertise.
269
In our early studies with some of my colleagues at Harvard and my teenagers at Warwick,
270
we found that like BCG consultants with less experience were worse at using AI.
271
And there's other studies that show less expertise results in worse AI outcomes, including philanthropic itself.
272
And so I think one of the dangers you have is you're like, oh, I need these, like,
273
you know, there's some group people who gets this.
274
Nobody gets this, right?
275
So as a senior leader, you actually have the best tools available because you have that expertise and judgment to make decisions about things.
276
And you probably have a thrive under surgery too.
277
If you've been around for 20 years or you've been through financial crisis,
278
you've been through wars, this is going to be a time of a lot of change.
279
That doesn't mean that you're inflexible because you're a senior leader.
280
In fact, often the best senior leaders who got where they are, are good at deciding how to make changes.
281
Have you seen companies or leaders that you know really successfully use AI in strategy development itself?
282
I've talked to C -level leaders at very large funds who've done this.
283
I've spoken to senior leaders in large financial services organizations, certainly in software.
284
Yes, there are people doing this all the time, right?
285
And in fact, I would argue you're probably making a mistake if you're not getting a second opinion from the AI.
286
Like the models are very smart right now.
287
It's just like if you're getting medical advice, absolutely see your doctor.
288
But every piece of evidence is you should absolutely be feeding that into the AI system, too, right?
289
I would do the same thing with legal advice.
290
Listen to your lawyer first, but why not get a second opinion?
291
Second opinions are cheap right now.
292
And the best models, and again, I can't emphasize enough, you need to use one of the frontier models, which at the time of the recording,
293
that means, you know, Claude Fable 5 .1 of whatever it'll be in the next couple of days and GPD 6, right?
294
And, you know, maybe Google comes up with the model again.
295
But you need to use one of these frontier models that you have to pay for.
296
And you should be asking a question like the only way you're going to find out what this is good
297
or bad is by using it
298
so by all means use
299
it for everything during the day take its opinions decide where it's dumb
300
that's a useful set of information for you yeah
301
when it comes to strategy does ai do you believe it ultimately favors large companies with more resources
302
and data um and those experienced leaders that you were talking about
303
or you know does it allow much smaller companies to compete compete, you know, with capabilities that they couldn't access before.
304
Both of those things.
305
One of the most depressing things I see
306
when I talk to boards of very large companies
307
and to C -level leaders is a lot of them have come to believe they just can't change, right?
308
So I show them what AI can do and they're like, well, I guess we'll ride this out.
309
And I'm like, that's an insane answer.
310
You have 100 ,000 people working for you.
311
The smartest people in the world are working for your company.
312
You have customers and resources.
313
The idea that you can't be flexible and compete seems to be overly baked in right now.
314
And like I'm a business school professor, I understand we talk a lot about,
315
you know, how difficult it is to survive disruption and how much inertia companies have.
316
But it's also a very weird decision because you have all the resources
317
because who's going to figure out how to use AI in your organization is going to be people in your company.
318
Experience has returns, right?
319
The startup doesn't have any of the same barriers you do, but they don't have the experience anymore.
320
And that experience matters.
321
It's no longer a 20 -something knows everything about this.
322
The more experience you have, the more you know about AI in your field.
323
And everyone's using the same AI models.
324
It's not like the startup you're talking to is going to have a better model than, you know, than Claude.
325
So I worry a little bit that people are sort of giving up.
326
On the startup side, it's never a better time to be a startup company.
327
Startups are all about that you're the top 0 .1 % at something
328
and everything else you're terrible at and you hope that doesn't destroy your company.
329
Now we have a tool that makes you good at everything you were good at.
330
Now it makes that stuff all feel a little bit similar to each other, but that's okay.
331
You know, I'm somebody who, when I was in my first startup company, we had to do payroll
332
and I didn't realize you could pay a payroll company a couple cents per payroll period to do taxes for you.
333
So I would do taxes by hand that excel for every person of the company.
334
It took like eight hours every week before I realized the better one.
335
Like AI will just tell me I didn't have to do that.
336
It could take care of it for me.
337
There's so much that it could help solve.
338
So it's accelerating everybody, right?
339
And we haven't talked enough about the individuals in your company.
340
They're also accelerating.
341
The question is, what do you do with that?
342
I have to ask about sort of all the fear mongering
343
now that's going on about sort of unleashing agents in your organization
344
and having them create chaos or start doing whatever they want to, you know, without listening to what leaders want.
345
So tell me your reaction to sort of all the recent news.
346
with OpenAI and Anthropa calling for maybe a slowdown in AI development?
347
So I think we need to differentiate between future model capabilities, the un -guard railed version of models that are being tested now, and current capabilities.
348
So I think there is enough smart people worried about future model development, and we're still in this exponential curve,
349
that if they're worried about a slowdown, government action or other action to pace the frontier
350
and figure out what the right thing to do is and give people time,
351
like, so that we are in the loop as deciding what happens seems reasonable, right?
352
What form that takes is up for grabs.
353
How important, how risky it is really is kind of up for grabs.
354
But like, there's no reason not to have some, you know, to have policy to prevent existential risk.
355
I think we're in for a cybersecurity environment that is apocalyptic in the near term.
356
You're going to have, you know, it might be solvable.
357
Defense could beat offense.
358
But I think we are in a world where when we get unguard railed models, cybersecurity is going to be a real issue.
359
Within organizations, using guard railed models, there just isn't the same kind of risk.
360
So I think we need to differentiate between those things.
361
But how should organizational leaders be thinking about that environment, which you said might become apocalyptic for,
362
you know, rogue agents from outside thwarting their competitive strategy because they're stealing information or causing disruption?
363
I think that you should think about it not so much as your competitors unleashing things on you to cause trouble.
364
I think it's going to be that you're going to have to deal with security environments where agents,
365
for many, many reasons, are going to be attempting to penetrate your security.
366
I think that when I talk to CSIOs,
367
I mean, OpenAI pulled 20 % of their staff off of research and 20 % of the resources to hardening their systems.
368
I have talked to leaders of large financial institutions are pulling 50 % plus of their staff and try, everyone's trying to figure what to do.
369
At the same time, defense has increased.
370
The amount of patching you need to do has increased, which also creates problems.
371
So like, we don't know how this goes through,
372
but I think we're going to have a very disruptive period where the cybersecurity environment is going to get kind of intense.
373
And that will look very, you know, like crazy from the outside.
374
I think that thinking about how you're going to deal with that now is really important.
375
I think you're probably underestimating how difficult that situation is going to be based on the experts I talk to.
376
I'm not a cybersecurity expert.
377
So I don't think it's
378
so much attackers bothering
379
you as much as every part of your system exposed to the internet is going to be a target for agents.
380
Maybe accomplishing completely random benign tasks, right?
381
If I ask an agent swarm to price products, it might decide to really get good pricing for Christmas because you wanted to buy a sweater.
382
I really need to get inside of this wool mills systems
383
because that's the only way can find out the pricing I need and relentlessly attack that will mill.
384
That wasn't a malicious attack.
385
That was a, I just need to find the price per wool that you have in the futures contract market.
386
And I've become obsessed with this.
387
So it's going to get very complicated.
388
We're dealing with kind of alien technology to some extent.
389
What's the biggest strategic mistake
390
that you see companies making right now as they're trying to figure out what AI means for their business?
391
So either, there's two giant ones.
392
One of them is fixing on the ability of AI right now.
393
I have this list I have of what the hot thing is in AI, and every three months it changes it.
394
Everyone goes all in.
395
All of these things are because we're anchoring on what the capabilities of AI are today.
396
So the hot thing that has been recently is everyone's spending too much on tokens.
397
Well, guess what?
398
Once you start thinking about multiplayer AI, where many people in your organization are using it, token costs become more reasonable.
399
All the AI companies are now reducing the costs of models
400
and having models use dumber agents to do work that are cheaper.
401
Everyone who went all in on cost was the main issue.
402
That may or may not be the concern a few months from now.
403
I mean, I can't tell you.
404
It depends on your organization and the choices you're making.
405
So I think there's a danger in fixating on what you have now
406
and satisficing based on the systems and capabilities you have today and not realizing that we're still in the exponential.
407
That is by far the number one problem, right?
408
The second problem is trusting that other people have got this nailed down.
409
Like every vendor you talk to is also making this up as we go along.
410
Every consultant you talk to does not have 17 successful cases.
411
And if they have successful cases, it's for GPT -4, a model that's obsolete and no one would use for anything anymore.
412
And your phone's model is more powerful than that.
413
So you have to build the capacity to make your own solutions in -house.
414
That doesn't mean write all your code or anything else yourself, but it does mean build that capacity internally.
415
So let's talk about that first problem or mistake that you see.
416
How do you teach yourself to embrace the idea of exponential advancement
417
and be ready sort of for where the puck is going?
418
So to take an idea from Harvard, from Mike Tushman and company, it was the idea that you would need ambidexterity.
419
You need to be able to do two things at once.
420
You need to be able to exploit what's happening today, but realize you're still exploring.
421
That's why the lab is so important.
422
Without your own internal flexible team that's building impossible things for the future, you get stuck.
423
so there's a few questions i ask organizations whenever
424
i talk to them the first is what have you stopped
425
doing as a result of ai ai has made something in your organization trivial what is the trivial thing
426
that used to be really important
427
but it's no longer important you have to stop doing
428
it ai just does it second thing is what are you building
429
it's impossible you should be building at least one thing
430
that is unbelievable and that would change the industry
431
if you're not building that you're not being ambitious enough even more
432
so you should be building something doesn't work yet because you're betting
433
that exponential curve will keep growing and at some point it will work.
434
And then the fourth question is, do you have leadership lab and crowd in place?
435
So I think if you have those four questions answered, you're in much better shape for being ambidextrous.
436
Yeah.
437
And how do you staff your entire organization with the talent that's able to execute on everything you're talking about?
438
It's also a really hard question.
439
Everything from psychological safety to how do we build the right structure?
440
All of that now is you're cashing all of that in.
441
If you have a good structure and a good organization and a good culture and the right reward systems,
442
people in your organization are discovering use cases and they're telling you about it.
443
If you're not, then everyone's hiding all their uses
444
and nothing you could do can change that because they're scared
445
if they tell you they'll get fired or they won't get respected, they won't get rewarded.
446
And why would they ever tell you about how they're using these systems?
447
So you need expertise in your organization, finding this and people willing to talk about that.
448
And that comes from having a good culture, structure, reward system, openness, good management.
449
Those are more urgent than ever.
450
You talked about sort of the first questions that you ask leaders
451
when you're beginning to advise them on how they should be managing this.
452
As you sort of get down the line and they sort of face challenges, what are sort of the most common hurdles
453
you see to prevent them from sort of fully moving forward into this new world of sort of AI led strategy?
454
I mean, I think that your danger is in a lot of organizations making this a normal technology, right?
455
Like there's a desperate desire to make this like any other technology
456
that we can swallow and we'll turn into a project that has X dollars associated with it.
457
And we have a six month RP process
458
and then an eight month discovery process that like that is not going to work.
459
You're going to experiment.
460
You're going to have to fail.
461
Like nobody has answers your question here.
462
So if you try and do this as a completely de -risked thing,
463
you end up in a position where you end up replicating other large technology projects.
464
And that is going to be obsolete the moment you launch it.
465
And you haven't built the dynamic capability to change what you change.
466
Making sure that you stay hungry for the future is the real problem here.
467
And keep up with that exponential.
468
And I only have so much good advice.
469
It's a hard problem to solve.
470
Yeah, absolutely.
471
So it's not sort of an AI adoption or even transformation plan.
472
it's really revolutionizing the way you think about what your business is designed to do
473
the one thing i've gotten most wrong about ai in the
474
last year is i thought agents would be harder to pull off than they are i thought
475
you would to organize teams of agents i thought we need to use all of our organizational design knowledge
476
and it turns out we don't that there are 100 000 agent swarms
477
that are self -organizing i didn't expect that to happen
478
but it makes sense when you think about the reason we have organizations and they're complicated is People have limited knowledge.
479
People have limited, have different interests in mind.
480
They have conflicts.
481
They don't have the same goal.
482
AI agents share all those things.
483
So suddenly coordination is not as hard as we thought the problem was going to be.
484
Like coordination turns out to be an easier problem than we thought.
485
And as a result of all of that, right?
486
Like suddenly you can get things done with AI in the way you'd expect.
487
So the problem becomes now in the leadership side, what do I want the organization to do?
488
And if your KPIs and other material are all baked around the idea of human outputs, You don't want 100 times more PowerPoint.
489
You don't want 100 times more code.
490
You want different outputs that are better.
491
So you need to sit back and think about what those outputs
492
and what those goals are and be aiming to maximize those goals.
493
Otherwise, you have internal things and you end up just producing 100 times more code, 100 times more PowerPoint.
494
I mean, the one thing, if there's one lesson I learned from my MBA program before I got my PhD,
495
it's that people do what they're incentivized to do.
496
And you need to think really hard about the incentives in your organization.
497
So the incentive to your organization are minimize the impact of AI,
498
use AI as a productivity tool to do more of what my KPI tells me I should do.
499
It's not transformation.
500
Let's zoom out and talk about a few bigger picture issues.
501
First, you know, obviously there's a fear that this will lead to massive unemployment.
502
So what do you see as sort of the future labor market in this new AI -led world?
503
That is the biggest question and debate among economists about this topic and we don't have easy answers, right?
504
So you can easily theorize that like the junior market gets squeezed because junior employees don't have the expertise.
505
Why use them when you can use AI?
506
And there's some indications of that, but there's also some indications there may be more hiring because I mean,
507
you know, now you can use coders to do things that organizations didn't have coders now can write their own code.
508
So you start insourcing and now everybody's doing their own marketing firm.
509
So maybe you lose jobs at like large centralized firms, but inside the organization, jobs grow because there's more demand it's really hard to know right?
510
You can imagine, and the AI labs, to be clear, are aiming for the jobpocalypse.
511
They want AI to replace all human labor.
512
That is their explicit goal.
513
I think there's reasons that, A, they might be wrong about this, and B, we don't want them to win,
514
so we shouldn't let that be the future that we get.
515
But that is their goal.
516
So I don't know what the outcome is in the long term in terms of employment per se,
517
what the impact on jobs is going to be, and nobody does.
518
So we're watching that closely, and everything is very confused right now in terms of the indicators
519
we don't really have an answer as a technology advances so quickly
520
and we don't even know what's going to happen with it as you said
521
you know sort of you have to be developing your strategy dynamically
522
you have to be prepared for change you have to experiment
523
and fail you have to be ready for exponential advancements how do we get better at predicting not just the opportunities,
524
but also the really terrifying risks?
525
I mean, things are accelerating quite quickly.
526
I think it is hard to bounce those things out.
527
I think that this is a general purpose technology.
528
It's many different things at once.
529
There are existential risks associated with it potentially.
530
There are certainly near -term risks.
531
I worry a lot about, again, automation of jobs instead of augmentation of jobs.
532
I worry about the people being taken out of all the interesting parts of their job
533
and their only job is verifying the results of AI content and be fired if it's wrong.
534
It seems like a terrible way to live.
535
I worry about impacts on our information environment.
536
I mean, at the same time, I'm worried about as an instructor that people are cheating with AI and then they don't learn as much.
537
At the same time, we know AI is also an incredibly good tutor,
538
so it can transform teaching for the positive just like it can for the negative.
539
I think that it's a great freeing aspect for people who are stuck, that they can get more stuff done before.
540
We certainly know it gives good medical advice, legal advice, things like that.
541
Still consult your doctor and lawyer.
542
I'm not trying to tell you not to do that.
543
But I think that it's a mixed bag.
544
So how do we react to that?
545
We react with policy to encourage good uses and mitigate bad ones.
546
We encourage it by being meaningful about how we're using these systems and not doing them blindly.
547
We do that by as leaders making choices about how these systems are used
548
and working in concert with our team and our organization to make sure that we're not disempowering people on math.
549
And I think good leadership will start to show through in these kinds of circumstances.
550
Well, Ethan, thank you so much for wrestling with all these really tricky issues with us.
551
I've learned a lot about AI from you over the years, and especially a lot from your new book.
552
Thanks so much for being with me today.
553
Thank you for having me.
Télécharger l'application
Tout pour parler couramment
Prononciation IAChaque phrase notée
Phonétique (IPA)Maîtrisez chaque son
VocabulaireEnrichissez votre lexique
Jeu de vocabulaireApprenez en jouant
Même chaîne
C11
Leading with Courage in an Age of Uncertainty
Daily ConversationUniversity StudentsWorking Adults
B21
How to Find a Mentor Who Can Accelerate Your Career | Christine vs. Work
Business EnglishInterview EnglishUniversity Students
C17
Turn Anxiety into Action
Business EnglishWorking Adults
B2Stand Out in a Job Interview | The Harvard Business Review Guide
Business EnglishInterview EnglishUniversity Students
C15
How to Be Creative on Demand
Business EnglishUniversity StudentsWorking Adults
C15
Commander's Intent Enables Great Leadership
Business EnglishWorking Adults
C11
Leisure's Not a Luxury. It's a Requirement for Top Leaders
Business EnglishUniversity StudentsWorking Adults
C13
The Five Competitive Forces That Shape Strategy
Business EnglishUniversity StudentsWorking Adults
B22
How to Get Good at Small Talk, and Even Enjoy It
Business EnglishDaily ConversationUniversity Students
C13
What Makes a Great Leader?
Daily ConversationUniversity StudentsWorking Adults
C14
Can Work Make You Happy? Should It?
Business EnglishUniversity StudentsWorking Adults
C13
Use Strategic Thinking to Create the Life You Want
Business EnglishUniversity StudentsWorking Adults
C142
You Need to Be Bored. Here's Why.
Business EnglishUniversity StudentsWorking Adults
C16
A Smarter Way to Delegate: What to Keep vs. What to Hand Off
Business EnglishWorking Adults
C13
Identity Crisis: Why Defining Yourself by Your Career Is a Problem
Business EnglishUniversity StudentsWorking Adults
C13
Lean Into Imposter Syndrome, Don't Give In to It
Business EnglishUniversity StudentsWorking Adults
C14
If Your Team Never Fails, It Isn’t Learning
Business EnglishUniversity StudentsWorking Adults
C14
A Plan Is Not a Strategy
Business EnglishUniversity StudentsWorking Adults
C1Why Great Teams Don't Need to Like Each Other
Daily ConversationUniversity StudentsWorking Adults
C1Masterclass: Match Your Pace to the Moment
Daily ConversationUniversity StudentsWorking Adults
FOR YOU
Vidéo recommandée
C11
828 phrases
Harvard AI Professor: The People Most at Risk From AI Aren’t Who You Think
Daily ConversationUniversity StudentsWorking Adults
C11
114 phrases
Harnessing generative AI in architecture, engineering & construction
Daily ConversationUniversity StudentsWorking Adults
C14
75 phrases
Become a great strategic thinker | Ian Bremmer
Daily ConversationUniversity StudentsWorking Adults
C17 phrases
How AI Empowers Tech Leaders Siwei Shen
Daily ConversationUniversity StudentsWorking Adults
C12
74 phrases
How AI is helping brands identify consumer trends faster
Daily ConversationUniversity StudentsWorking Adults
C11
293 phrases
Is AI Killing Human Fashion Design? Who Will Shape the Future of Design? Fashion Tea #6
Daily ConversationUniversity StudentsWorking Adults
C184 phrases
The 4-Hour Workweek in a World of AI
Daily ConversationUniversity StudentsWorking Adults
C11
623 phrases
The Only SaaS Moats Left in 2026
Daily ConversationUniversity StudentsWorking Adults
C11
995 phrases
Wharton’s Top AI Professor: Stop Starting Software Companies — The Real AI Money Is Here
Daily ConversationUniversity StudentsWorking Adults
C15
182 phrases
How to Become an AI Product Manager in 2026 | Ex-Google, Microsoft
Daily ConversationUniversity StudentsWorking Adults
C11
526 phrases
What AI Can Never Replace w/ Megan Kobzej | Cloud English Podcast Episode 91
Daily ConversationUniversity StudentsWorking Adults
C14
36 phrases
The Most Powerful Leadership Question | Simon Sinek
Daily ConversationUniversity StudentsWorking Adults
Vocabulaire et conseils d’expression pour cette leçon
Cette leçon d’expression orale s’appuie sur la vidéo « How AI Is Changing Strategy—and 4 Skills Leaders Need Next ». Les mots qui reviennent le plus souvent : organization, leader, Ethan, Mollick. Cette vidéo contient 483 phrases et 6952 mots à répéter en shadowing. La partie parlée dure 32:32. Le locuteur parle vite, environ 214 mots par minute : attendez-vous à des liaisons et à des sons réduits. 84 % des mots font partie des 3 000 mots les plus courants en anglais ; le reste mérite d’être vu avant de commencer.
Vocabulaire clé de cette vidéo
15 mots de la vidéo à retenir, avec leur prononciation et leur sens :
| Mot | Prononciation | Sens |
|---|---|---|
| expertise nom | /ˌɛkspɜːˈtiːz/ | compétence, expertise |
| capability nom | /ˌkeɪ.pəˈbɪl.ɪ.ti/ | capacité |
| strategic adjectif | /stɹəˈti.dʒɪk/ | stratégique |
| productivity nom | /ˌpɹɒdʌkˈtɪvəti/ | productivité |
| accomplish verbe | /əˈkɑm.plɪʃ/ | accomplir |
| judgment nom | /ˈd͡ʒʌd͡ʒ.mənt/ | jugement |
| organizational adjectif | /ˌɔː(ɹ)ɡənɪˈzeɪʃənəl/ | organisationnel |
| startup nom | /ˈstɑɹtˌʌp/ | mise en route, mise en train |
| automation nom | /ˌɔ.təˈmeɪ.ʃən/ | automatisation |
| output nom | /ˈaʊtpʊt/ | sortie |
| experiment nom | /ɪkˈspɛɹ.ə.mənt/ | expérience |
| loop nom | /luːp/ | boucle |
| urgent adjectif | /ˈɜː.d͡ʒənt/ | urgent |
| frontier nom | /ˈfɹʌntɪə/ | frontière |
| extent nom | /ɪkˈstɛnt/ | mesure |
Les verbes à particule que vous entendrez
| Mot | Sens |
|---|---|
| figure out verbe | cerner |
| come up with verbe | inventer, imaginer |
| end up verbe | se terminer, se finir |
| find out verbe | découvrir, savoir |
| go off verbe | exploser |
| eat up verbe | dévorer |
| fall behind verbe | être en retard |
| give up verbe | se rendre |
Des phrases à répéter
Des phrases courtes et complètes de la vidéo, à réutiliser dans la conversation de tous les jours :
- What do you think companies should be doing now?
- What's the product you're trying to produce?
- What do you do about that?
- How do we incentivize people?
- Second thing is, what are you building that's impossible?
La grammaire de cette vidéo
Les structures que le locuteur utilise le plus, avec les mots exacts de la vidéo :
| Structure | Dans la vidéo |
|---|---|
| Present perfect have/has + participe passé — une action passée qui compte encore maintenant | have come · you've been · has become |
| Voix passive be + participe passé — l’accent est mis sur ce qui arrive, pas sur qui le fait | be overly baked · is used · is done |
| Propositions relatives who / which + proposition — une précision sur une personne ou une chose | organization, which is · model, which is · people who are |
Prononciation à surveiller
Le locuteur utilise 186 contractions et formes réduites, comme you're, don't, they're. Prononcez-les sous leur forme courte, telles que vous les entendez.
- Les sons « sh » et « zh »: accomplish /əˈkɑm.plɪʃ/, organizational /ˌɔː(ɹ)ɡənɪˈzeɪʃənəl/, automation /ˌɔ.təˈmeɪ.ʃən/, transformation /ˌtɹæns.fɔɹˈmeɪ.ʃən/, differentiate /dɪf.əˈɹɛn.ʃi.eɪt/
- Mots longs — placez bien l’accent: capability /ˌkeɪ.pəˈbɪl.ɪ.ti/, productivity /ˌpɹɒdʌkˈtɪvəti/, organizational /ˌɔː(ɹ)ɡənɪˈzeɪʃənəl/, automation /ˌɔ.təˈmeɪ.ʃən/, experiment /ɪkˈspɛɹ.ə.mənt/
Les sons difficiles pour les francophones :
- /h/ — il se souffle, il n’est pas muet: hire /ˈhaɪɹ/, hungry /ˈhʌŋ.ɡɹi/, behalf /bɪˈhæf/
- /tʃ/ et /dʒ/ — à ne pas adoucir en « ch » et « j »: strategic /stɹəˈti.dʒɪk/, judgment /ˈd͡ʒʌd͡ʒ.mənt/, urgent /ˈɜː.d͡ʒənt/, encourage /ɪnˈkʌɹ.ɪd͡ʒ/, engage /ɪnˈɡeɪd͡ʒ/
- /r/ anglais — langue recourbée, sans frotter la gorge: strategic /stɹəˈti.dʒɪk/, productivity /ˌpɹɒdʌkˈtɪvəti/, organizational /ˌɔː(ɹ)ɡənɪˈzeɪʃənəl/, startup /ˈstɑɹtˌʌp/, experiment /ɪkˈspɛɹ.ə.mənt/
Comment s’entraîner avec cette vidéo
- Écoutez la vidéo en entier une fois sans parler et notez les mots que vous ne connaissez pas.
- Commencez à la vitesse 0,75×, répétez phrase par phrase, puis revenez à la vitesse normale quand cela devient facile.
- Enregistrez-vous et comparez avec l’original, en faisant attention à des mots comme expertise, capability, strategic.
Qu'est-ce que la technique du Shadowing ?
Le Shadowing est une technique d'apprentissage des langues fondée sur la science, développée à l'origine pour la formation des interprètes professionnels. Le principe est simple mais puissant : vous écoutez de l'anglais natif et le répétez immédiatement à voix haute — comme une ombre suivant le locuteur avec un décalage de 1 à 2 secondes. Les recherches montrent une amélioration significative de la précision de la prononciation, de l'intonation, du rythme, des liaisons, de la compréhension orale et de la fluidité.