تدريب Shadowing: Elon Musk & Rishi Sunak Interview IN FULL - تعلم التحدث بالإنجليزية عبر الفيديو
اختر وضع التدريب
كل وضع يفتح غرفة تدريب بملء الشاشة.
702 جمل
إذا كانت الجمل قصيرة أو طويلة جدًا، انقر على Edit لتعديلها.
1
okay all right well good evening everybody welcome Elon thanks for
2
being here thank you for having me we feel we feel very privileged we're excited to have you right
3
so I'm going to start with some questions
4
and then we're going to open it up let me get straight into it so
5
Bill Gates said there is no one in our time who
6
has done more to push the bounds of science innovation than
7
you well that's it that's a nice thing to have anyone say about you nice coming from bill gates
8
but oddly enough
9
when it comes to ai actually for around a decade you've almost been doing the opposite
10
and saying hang on you need to think about what we're doing
11
and what we're pushing here and what do we do to make this safe and
12
and actually maybe we shouldn't be pushing as faster as hard
13
as we are like i mean you've been doing it for a decade like what was it
14
that caused you to think about it that way
15
and you know why do we need to be worried yeah
16
i've been I've been somewhat of a Cassandra for quite a while, where people would, I would tell people, like, we should really be concerned about AI.
17
They'd be like, what are you talking about?
18
Like, I've never really had any experience with AI.
19
But since I was immersed in technology, I have been immersed in technology for a long time, I could see it coming.
20
So, but I think this year was, there have been a number of breakthroughs.
21
You know, the point at which someone can see a dynamically created video of themselves,
22
You know, like, somebody could make a video of you saying anything in real time or me.
23
And so there's sort of the deep -baked videos, which are really incredibly good.
24
In fact, sometimes more convincing than real ones.
25
Deep real.
26
And then, obviously, things like ChatGPT were quite remarkable.
27
I saw GPT -1, GPT -2, GPT -3, GPT -4, the whole sort of lead up to that.
28
So it was easy for me to kind of see where it's going.
29
If you just sort of extrapolate the points on a curve and assume that trend will continue,
30
then we will have profound artificial intelligence and obviously at a level that far exceeds human intelligence.
31
So I'm glad to see at this point
32
that people are taking safety seriously and I'd like to say thank you for holding this AI safety conference.
33
I think actually it will go down in history as being very important.
34
I think it's really quite profound.
35
And I do think overall that the potential for artificial intelligence, AI, to have most likely a positive effect
36
and to create a future of abundance where there is no scarcity of goods and services.
37
But it is somewhat of the magic genie problem, where if you have a magic genie that can grant all the wishes,
38
usually those stories don't end well.
39
Be careful what you wish for, including wishes.
40
So you talked a little bit about the summit, and thank you for being engaged in it, which has been great, and people enjoyed having you there.
41
Potter's been in this dialogue.
42
Now, one of the things that we achieved today in the meetings between the companies
43
and the leaders was an agreement that externally, ideally, governments should be doing safety testing of models before they're released.
44
I think this is something that you've spoken about a little bit.
45
It was something we worked really hard on because my job in government is to say, hang on, there is a potential risk here,
46
not a definite risk, but a potential risk of something that could be bad.
47
My job is to protect the country.
48
And we can only do that
49
if we develop the capability we need in our safety institute
50
and then go in and make sure we can test the models before they are released.
51
Delighted that that happened today, but what's your view on what we should be doing?
52
You've talked about the potential risk.
53
Again, we don't know.
54
But what are the types of things governments like ours should be doing to manage and mitigate against those risks?
55
Well, I generally think that it is good for government to play a role when the public safety is at risk.
56
So, really, for the vast majority of software, the public safety is not at risk.
57
I mean, if the app crashes on your phone or your laptop, it's not a massive catastrophe.
58
But when talking about digital superintelligence, I think, which does pose a risk to the public,
59
then there is a role for government to play to safeguard the interests of the public.
60
And this is, of course, true in many fields, you know, aviation, cars.
61
I deal with regulators throughout the world because of Starlink being communications,
62
rockets being aerospace, and cars being vehicle transport.
63
So I'm very familiar with dealing with regulators.
64
And I actually agree with the vast majority of regulations.
65
There's a few that I disagree with from time to time.
66
0 .1 % probably, or less than 1 % of regulations I disagree with.
67
So there is some concern from people in Silicon Valley who have never dealt with regulators before
68
and they think that this is going to just crush innovation and slow them down and be annoying.
69
And it will be annoying, it's true, they're not wrong about that.
70
But I think we've learned over the years that having a referee is a good thing.
71
And if you look at any sports game, there's always a referee.
72
And nobody's suggesting, I think, to have a sports game without one.
73
And I think that's the right way to think about this,
74
is for Govan to be a referee to make sure the sportsmanlike conduct and that the public safety is addressed,
75
that we care about public safety.
76
Because I think there might be, at times, too much optimism about technology.
77
And I say that as a technologist, so I ought to know.
78
And on balance, I think that the AI will be a forceful good,
79
most likely, but the probability of it going bad is not 0%.
80
Yeah.
81
So we just need to mitigate the downside potential.
82
And then how you talk about referee, and that's what we're trying to say.
83
Demis right there.
84
Yeah, well, there we go.
85
I mean, you know, we talk about this, and Demis and I discussed this a long time ago.
86
We're like literally facing right at it.
87
And actually, you know, Demis, to his credit and the credit of people in the industry, did say that to us.
88
I think, you know, Demis said it's not right that Demis and his colleagues are marking their own homework, right?
89
There needs to be someone independent, and that's why we've developed the Safety Institute here.
90
I mean, do you think governments can develop the expertise?
91
One of the things we need to do is say, hang on, you know, Demis, Sam, all the others have got a lot of very smart people doing this.
92
Governments need to quickly tool up capability -wise, personnel -wise, which is what we're doing.
93
I mean, do you think it is possible for governments to do that fast enough, given how quickly the technology is developing, or what do we need to do to make sure we do do it quick enough?
94
No, I think it's a great point you're making.
95
The pace of AI is faster than any technology I've seen in history by far.
96
And it seems to be growing in capability by at least five -fold, perhaps ten -fold per year.
97
It'll certainly grow by an order of magnitude next year.
98
Yeah.
99
So, and government isn't used to moving at that speed.
100
But I think even if there are not firm regulations, Even if there isn't an enforcement capability,
101
simply having insight and being able to highlight concerns to the public will be very powerful.
102
So even if that's all that's accomplished, I think that will be very, very good.
103
Okay.
104
Well, hopefully we can do better than that.
105
Hopefully, yeah.
106
No, but that's helpful.
107
Actually, we were talking before.
108
It was striking.
109
You're someone who spent their life in technology, living Moore's Law, and what was interesting over the last couple of days, he's talking to everyone who's doing the development of this,
110
and I think you concur with this, is just the pace of advancement here is unlike anything all of you have seen in your careers in technology.
111
Is that fair?
112
Because you've got these kind of compounding effects from the hardware and the data and the personnel.
113
Yeah.
114
I mean, the two, currently the two leading centers for AI development are the San Francisco Bay Area and the sort of London area.
115
And there are many other places where it's being done, but those are the two leading areas.
116
So I think if the United States and the UK and China are sort of aligned on safety,
117
that's all going to be a good thing.
118
Because that's where the leadership is generally.
119
Actually, you mentioned China there.
120
So I took a decision to invite China to summit over the last couple of days.
121
And it was not an easy decision.
122
A lot of people criticized me for it you know, my view is if you're going to try to have serious conversation, you need to.
123
But what were your thoughts?
124
You do business all around the world.
125
You just talked about it there.
126
Yeah.
127
Should we be engaging with them?
128
Can we trust them?
129
Is that the right thing to have done?
130
If China is not on board with AI safety, it's somewhat of a moot situation.
131
The single biggest objection that I get to any kind of AI regulation or sort of safety controls are, well, China's not going to do it,
132
and therefore they will just jump into lead and exceed us all
133
but but actually China is willing to participate in AI safety
134
and thank you for inviting them
135
and they you know I think I think we should thank China for attending
136
when I was
137
when I was in China earlier this year the my main
138
subject of discussion with this the leadership in China was AI safety
139
and saying that this is really something that they should care about.
140
And they took it seriously, and you are too, which is great.
141
And having them here, I think, was essential, really.
142
If they're not participants, it's pointless.
143
It's pointless, yeah.
144
And I think we were pleased.
145
They were engaged yesterday in the discussions and actually ended up signing the same communique that everyone else did, which is a good start, right?
146
And we need everyone to approach us in a similar way if we're going to have, I think, a realistic chance of resolving it.
147
I was going to tell you, you talked about innovation earlier and regulation being annoying.
148
There was a good debate today we had about open source, and I think you've kind of been a proponent of algorithmic transparency and making some of the X algorithms public.
149
And actually, we were talking about Jeffrey Hinton on the way in.
150
He's particularly been very concerned about open source models being used
151
by bad actors you've got a group of people who say they are critical to innovation happening in that distributed way.
152
Look, there's probably no perfect answer, and there is a tricky balance.
153
What are your thoughts on how we should approach this open source question, or where should we be targeting whatever regulatory or monitoring that we're going to do?
154
Well, the open source algorithms
155
and data tend to lag the closed source by six to 12 months
156
but so that so that but given the rate of improvement
157
that this actually therefore quite a big difference between the closed source and the and the open
158
if things are improving by a factor of let's say five
159
or more than being a year behind is your five times worse
160
so it's a pretty big difference and that might be actually an okay situation
161
but it's only will get to the point where you've about open source AI
162
that will start to approach human level intelligence or perhaps exceed it.
163
I don't know quite what to do about it.
164
I think it's somewhat inevitable that there will be some amount of open source, and I guess I would have a slight bias towards open source, because at least you can see what's going on.
165
Whereas closed source, you don't know what's going on.
166
Now, it should be said with AI that even if it's open source, do you actually know what's going on? because if you've got a gigantic data file and,
167
you know, sort of billions of data points or weights and parameters, you can't just read it and see what it's going to do.
168
It's a gigantic file of inscrutable numbers.
169
You can test it when you run it.
170
You can run a bunch of tests to see what it's going to do, but it's probabilistic as opposed to deterministic.
171
It's not like traditional programming where you've got very discrete logic
172
and the outcome is very predictable and you can read each line and see what each line is going to do.
173
A neural net is just a whole bunch of probabilities.
174
It sort of ends up being a giant comma -separated value file.
175
It's like, our digital god is a CSV file?
176
Really?
177
okay
178
that is kind of what it is that point you've just made is one
179
that we have been talking about a lot
180
because again conversations the people who are developing their technology make the point
181
that you've just made it is not like normal software where
182
there's predictability about inputs improving leading to this particular output improving
183
and as the models iterate
184
and improve we don't quite know what's going to come out the other end I think Demis would agree with that, which is why I think there is this bias
185
that we need to get in there while the training runs are being done, before the models are released to understand what is this new iteration brought about in terms of capability,
186
which it sounds like you would agree with.
187
I was going to shift gears a little bit.
188
You've talked a lot about human consciousness, human agency, which actually might strike people as strange, given that you are known for being such a brilliant innovator
189
and technologist but it's quite heartfelt when I hear you talk about it
190
and the importance of maintaining that agency in technology and preserving human consciousness.
191
Now, it kind of links to the thing I was going to ask is, when I do interviews or talk to people out and about in this job about AI, the thing that comes up most,
192
actually, is probably not so much the stuff we've been talking about, but jobs.
193
It's what does AI mean for my job?
194
Is it going to mean that I don't have a job or my kids are not going to have a job?
195
Now, my answer as a policymaker, as a leader, is actually AI is already creating jobs, and you can see that in the companies that are starting.
196
Also the way it's being used is a little bit more as a co -pilot, necessarily, versus replacing the person.
197
There's still human agency, but it's helping you do your job better, which is a good thing.
198
And as we've seen with technological revolutions in the past.
199
clearly there's change in the labor market the amount of jobs i was quoting an mit study today
200
that they did a couple of years ago something like 60 of the jobs at
201
that moment didn't exist 40 years ago so hard to predict
202
and my job is to create an incredible education system whether
203
it's at school whether it's retraining people at any point in their career
204
because ultimately if we've got a skilled population they'll be able to keep up with the pace of change
205
and have a good life but you know that it's still a concern
206
and you know you what would your kind of observation be on AI
207
and the impact on labor markets and people's jobs and how they should feel about that as they think about this?
208
Well, I think we are seeing the most disruptive force in history here.
209
You know, we have for the first time, we will have for the first time something that is smarter than the smartest human.
210
And that, I mean, it's hard to say exactly what that moment is, but there will come a point where no job is needed.
211
You can have a job if you want to have a job for sort of personal satisfaction, but the AI will be able to do everything.
212
So, I don't know if that makes people comfortable or uncomfortable.
213
You know, that's why I say if you wish for a magic genie, that gives you any wishes you you want, and there's no limit.
214
You don't have those three -wish limit, nonsense.
215
You just have as many wishes as you want.
216
So it's both good and bad.
217
One of the challenges in the future will be, how do we find meaning in life if you have a magic genie that can do everything you want?
218
I do think we, it's hard, when there's new technology, it tends to have, usually follow an S -curve.
219
In this case, we're going to be on the exponential portion of VS Code for a long time.
220
And you'll be able to ask for anything.
221
We won't have universal basic income.
222
We'll have universal high income.
223
So in some sense, it'll be somewhat of a leveler or an equalizer, because really, I think everyone will have access to this magic genie.
224
And you'll be able to ask any questions.
225
It'll certainly be good for education.
226
It'll be the best tutor and the most patient tutor.
227
there will be no shortage of goods
228
and services will be an age of abundance I think
229
if I'd recommend people read in banks the banks culture books
230
are played the best envisioning effect on probably they're definitely by
231
far the best envisioning of an AI future there's nothing even close.
232
So I'd recommend, really recommend Banks.
233
I'm a very big fan.
234
All these books are good.
235
Does not say which one, all of them.
236
So that'll give you a sense of what is a,
237
I guess, a fairly utopian or protopian future with AI.
238
Which is good from, as you said, it's a universal high income which is a nice phrase and that's it's good from a materialistic sense
239
abundance actually that it kind of then leads to the question
240
that you pose right I'm somebody who believes you know work gives you meaning a lot about
241
that as you know I think work is a good thing it you know gives people purpose in their lives
242
and if you then remove a large chunk of
243
that you know what does that mean and where do you get
244
that yeah you know where do you get that drive
245
that motivation that purpose I mean you're talking about it you you work a lot of hours
246
and you know as I was mentioning
247
when we were talking earlier I have to somewhat engage in deliberate suspension of disbelief
248
because when I'm putting so much blood sweat
249
and tears into a work project
250
and burning the you know 3 a .m oil then I'm
251
like wait why am I doing this I can just wait
252
for the AI to do it I'm just lashing myself for no reason yeah must be a glutton for
253
We call Demis and tell him to hurry up and then you can have a holiday, right?
254
It's a tricky thing because part of our job is to make sure
255
that we can navigate to that very largely positive place
256
that you're describing and help people through it between now
257
and then because these things bring a lot of change in the labor market as we've seen.
258
Yeah, I think it probably is generally a good thing
259
because there are a lot of jobs that are uncomfortable or dangerous or sort of tedious, and the computer will have no problem doing that.
260
They're happy to do that all day long.
261
So it's fun to cook food, but it's not that fun to wash the dishes.
262
But the computer's perfectly happy to wash the dishes.
263
I guess there is...
264
We still have...
265
sports like where humans compete in like the Olympics
266
and obviously a machine can can go faster than any human
267
but we still have we saw humans race against each other and
268
and have all you know have these sports competitions against each
269
other where even though the machines are better with visual I
270
guess competing to see who can be the best human at something
271
and people do find fulfillment in that
272
so I guess that's perhaps a good example of how even
273
when machines are faster than us stronger than us we still
274
find a way we still we still enjoy competing against other
275
humans to at least see who's the best human yeah that's a good that's a good analogy
276
and we've been talking a lot about managing the risks I just before we move on
277
and finish on AI is to talk a little bit about
278
the opportunities you're engaged in lots of different companies you're not being an obvious one
279
which is doing which is doing some exciting stuff I you touched on the thing
280
that I'm probably most excited about which is in education yeah
281
and i think many people will have seen sal khan's video
282
from earlier this year's ted talk about as you talked about
283
it's like personal tutor yeah an amazing personal tutor an amazing personal tutor
284
and we know the difference in learning having
285
that personalized tutor is incredible compared to classroom learning
286
so you can have every child have a personal tutor specifically for them
287
that then just evolves with them over time yeah
288
that could be extraordinary and so
289
that you know for me I look at I think gosh with
290
that is within reach at this point yeah
291
and that's one of the benefits I'm most excited about that
292
when you look at the landscape of things
293
that you see as possible what is it
294
that you know you are particularly excited about I think certainly
295
AI tutors are gonna be amazing that's already are I think there's also perhaps companionship
296
which may seem odd because how can the computer really be your friend
297
but if If you have an AI that has memory and remembers all of your interactions and has read everything, you're going to give it permission to read everything you've ever done,
298
so it really will know you better than anyone, perhaps even yourself, and where you can talk to it every day and those conversations build upon each other,
299
you will actually have a great friend.
300
As long as that friend can stay your friend and not get turned off or something.
301
Don't turn off my friends.
302
but I think that will actually be a real thing and
303
one of my sons has some learning disabilities has trouble making friends actually and I was like well
304
an AI friend would actually be great for him
305
that was a surprising answer that's actually worth reflecting on that's really interesting we're already seeing it actually as we deliver
306
psychotherapy anyway now doing far more by digitally and by telephone to people
307
and it's making a huge difference and you can see a world in
308
which actually you know AI can provide that social benefit to people
309
and just a quick question on on X
310
and then we should open it up to everybody you made a change
311
when you one of the many changes
312
but yeah one at one of the one of the chain love that letter yeah I've got a real thing about it.
313
You really do.
314
You really do.
315
One of the changes which kind of goes into the space that we have to operate in,
316
and this balance between free speech and moderation we grapple with as politicians, you were grappling with your own version of that.
317
And you moved away from a manual, human way of doing it, the moderation, to the community notes.
318
And I think it was an interesting change, right? it's not what everyone else has done it would be good you know what's what was the reasoning behind
319
that and why do you think that is a better way to do
320
that um yeah part of the problem is if you
321
if you empower people as sensors then well there's going to be some amount of bias they have um
322
and then whoever appoints the sensors is effectively in control of information
323
so then the idea behind community notes is well how do we have a consensus driven uh i mean
324
so it's not really censoring it
325
but consensus driven approach to truth uh how do we
326
or how do we how do we make things um the
327
least amount untrue like you can say like what you can't pass perhaps get to pure truth
328
but you can aspire to be more truthful um
329
so the the thing about community notes is it doesn't actually delete anything it simply adds context now
330
that context could be this thing is untrue for the following reasons.
331
But importantly, with community nodes, everything is open source, actually.
332
So you can see the software, every line of the software, you can see all of the data that went into a community node,
333
and you can independently create that community node.
334
So if you see manipulation of the data, you can actually highlight that and say, well, this...
335
that there appears to be some gaming of the system, and you can suggest improvements. So it's maximum transparency.
336
Transparency, yeah.
337
Which is, I think.
338
Combined with the kind of wisdom of the crowds and transparency to get to a better answer.
339
And really, one of the key elements of community notes is that in order for a note to be shown, people who have historically disagreed must agree.
340
And there is a bit of AI usage here.
341
So we'll populate a parameter space around each contributor to community notes.
342
And then a parameter space.
343
So everyone's got basically these vectors associated with them.
344
So it's not as simple as right or left.
345
It's saying it's several hundred vectors because things are more complicated than something right or left.
346
And then we'll do sort of inverse correlation say, like, okay, these people generally disagree, but they agree about this note.
347
Okay, so then that...
348
So then that gives the note credibility.
349
Okay.
350
Yeah, that's the core of it, and it's working quite well.
351
Yeah.
352
I've yet to see a note actually be present for more than a few hours that is incorrect.
353
So the batting average is extremely good.
354
And when I ask people, they'll say, oh, they're worried about community notes sort of being disinformation, like send me one.
355
And then they can't.
356
so I think it's quite good the general aspiration with the X platform is to inform
357
and entertain the public and to be as accurate as possible and as truthful as possible
358
even if someone doesn't like the truth people don't always like the truth not always
359
but that's the aspiration and I think
360
if we stay true to the truth then I think we'll find
361
that people use the system to learn what is going on and to I think actually truth pays.
362
So I think it'll be what, I mean, assuming you don't want to engage in self -delusion, then I think it's the smart move.
363
Excellent.
364
Very helpful.
365
Right.
366
Let's open it up to all our guests here.
367
We've got some microphones.
368
They'll come, put your hands up.
369
They'll come and find you we've got yes.
370
Go for it.
371
Thank you.
372
Good evening.
373
Alice Bentink from Entrepreneur First.
374
Thank you for a fascinating conversation.
375
I suppose a question for each of you.
376
Prime Minister, the UK has some of the best universities in the world.
377
We have the talent.
378
What will it take for the UK to be a real breeding ground for unicorn companies?
379
And Elon, being a founder in the UK is still a non -obvious career choice for the most exceptional technical talent.
380
What are the cultural elements that we need to put into place to change this?
381
Thank you both.
382
You want to go first?
383
Go for it.
384
Sure.
385
Well, you're right that there are cultural elements where the culture should celebrate creating new companies.
386
And there should be a bias towards supporting small companies because they're the ones that need nurturing.
387
The larger companies really don't need nurturing.
388
So just think of it sort of like a garden.
389
If it's a little sprout, it needs nurturing.
390
If it's a mighty oak, it doesn't need quite as much.
391
So I think that is a mindset change that is important.
392
But I should mention that London is,
393
you know, London and San Francisco or the Bay Area are really the two centers for AI.
394
So London is actually doing very well on that front.
395
I'd say the two leading locations on Earth.
396
San Francisco's probably ahead of London, but London's really very strong.
397
But London area.
398
Greater London.
399
Home county, I guess.
400
Keep going, keep going.
401
So I'm just saying objectively this is the case.
402
But you do need that to, you need the infrastructure, you need landlords who are willing to rent to new companies,
403
you need law firms and accountants that are willing to support new companies, and it's generally a mindset, it is a mindset change, and I think some of that is happening,
404
but I think really it's just culturally people need to decide this is a good thing.
405
Yeah, actually, well, thanks for what you said about the UK, it's something that we work hard on,
406
lots of people in the room are part of what makes this a fabulous place for of companies, including Alice.
407
So Alice, what I'd say is my job is to get all the nuts and bolts right, make sure that all of you starting companies can raise the capital that you need,
408
everything from your seed funding with our incredible EIS tax reliefs all the way through to your late -stage rounds, and we need reform of our pension funds,
409
and the Chancellor's got a bunch of incredible reforms to unlock.
410
capital from all the people who have it
411
and deploy it into growth equity right
412
that is a work in progress we're not there yet
413
but i think we're we're making we're making good progress we need talent we need people you know
414
so that means an education system that prioritizes the things
415
that matter and you've seen my reforms i go on about more maths more maths more maths um
416
but i think that is important but also attracting the best
417
and the brightest here if you look at our fastest growing companies in in this country
418
and i think it's probably the same in in the us
419
over half of them have a non -british founder right and
420
so that tells you we've got to be a place
421
that is open to the world's best and brightest entrepreneurial talent
422
so the visa regime that we've put in place i think does
423
that makes it easy for those people to come here
424
and then actually it's the thing
425
that we spent the beginning of the session talking about the regulation right making sure
426
that we've got a regulatory system that's pro -innovation
427
that yeah we've of course we always need guard rails on the things
428
that will worry us but we've got to create a space for people to innovate
429
and do different things you know those are all my jobs the thing
430
that is tougher is the thing that elon talked about
431
which is culture right it's how do you transpose
432
that culture from places like silicon valley across the world where
433
people are unafraid to give up the security of a regular
434
paycheck to go and start something and be comfortable with failure you you talk about
435
that a lot i think you talked about it more in
436
when you were playing games right but that you've got to be comfortable failing
437
and knowing that that's just part of the process and that is a It's a tricky cultural thing to do overnight, but it's an important part of, I think, creating that kind of environment.
438
Yeah, if you don't succeed with your first startup, it shouldn't be sort of a catastrophic career -ending thing.
439
It should be, you know, well, good.
440
I think generally it should be like, well, you gave it a good shot, you know, and now try again.
441
Exactly, yeah, and it's...
442
So, one thing I was going to mention is like, obviously creating a company is sort of a high -risk,
443
high -reward situation, but, and I don't know quite with how it works in the UK.
444
I think it's probably better than continental Europe.
445
But the stock options are very difficult in most parts of Europe.
446
I'm not sure how it is in the UK.
447
But if somebody's basically going to risk their life savings and the vast majority of startups fail,
448
so, I mean, you hear about the startups that succeed, but most companies are,
449
most startups consist of you know a massive amount of work followed by failure that's actually most most companies
450
and so it's a high risk higher award and and
451
so the higher award part does need to be there for it to make sense yeah i think
452
that was a very soft pitch for a tax policy
453
that i mean chancellor but actually i could tell you
454
so look a i agree and we have
455
so we have i think relative to certainly european countries
456
but certainly the us definitely california a much lower rate of capital gains tax okay right
457
so for those people who are risking
458
and growing something like we think the reward should be there at the end
459
so it's 20 capital gains tax rate um
460
and on stock options i don't know
461
if we've got anyone from index ventures in in the room
462
so you know index one of our leading vc funds here
463
okay they they do a regular report looking at most countries tax treatment of stock options yeah
464
and you know when i was chancellor you know treasury secretary equivalent you know We were pretty good, but we were fourth or fifth,
465
and I said, for exactly the reason that you mentioned, this has got to be the best place for innovators, we need to move that up.
466
And I think in the last iteration of that report, because of the changes that Jeremy and I have made, we have moved up to, I think, second from memory.
467
So hopefully that should give you and everyone else some comfort that we recognize that's important.
468
Because when people work hard and risk things, they should be able to enjoy the rewards.
469
High risk, high reward.
470
Yeah, and I think we very much have a tax system that supports that, and those are the values that, you know, I believe in, and I think most of us in this room probably do as well.
471
Right, next question.
472
I've got Seb in front of me, and then I'll come over here.
473
Go on, Seb.
474
Thanks very much.
475
We've talked about some really big ideas, global changing ideas.
476
I'm really interested, particularly in the context of creation of science and technology, super hubs and so on, How does that map onto the everyday lives of people living in,
477
say, Austin, Texas, to choose one around general, or in my case, Nottingham, East Midlands?
478
How do you see that evolving for people, you know, every day?
479
The sort of everyday effects of AI?
480
For context, Elon, so Seb runs our equivalent of CVS, right, or Walgreens.
481
And so, you know, actually I visited, right?
482
So he's got millions of people coming in his shops every day and it's making sure how do we make this relevant?
483
I think that's your question.
484
How is this relevant to that person?
485
You know, actually, I'll go first on that because I think it's a fair point.
486
I was just going over with the team a couple of things that we're doing because I was saying, like, well, how are we doing AI right now that it's making a difference to people's lives?
487
and we have this thing called gov .uk which is
488
which especially when we
489
when it happened several years ago was a pioneering thing all
490
the government information brought together on one website gov .uk and
491
so you need to get a driving license passport any interaction with government it was centralized
492
in a very easy relatively easy to use way uh better than most better than most yeah
493
so we are about so we're about to deploy ai across
494
that platform so that is something
495
that i think you know several million people a day use right
496
so a large chunk of the population is interacting with gov
497
.uk every single day to do all these day to day
498
tasks right every one of your customers is doing all those things
499
and so we're about to deploy AI into that to make
500
that whole process so much easier
501
because you know some people will be like look well I'm currently here
502
and I've lost my passport and my flights in five hours at the moment
503
that would require you know how many steps to figure out what you do yeah actually
504
when we deploy the ai it should be
505
that you could just literally say that
506
and boom boom boom boom this is what we're going to do walk you through it
507
and that's going to benefit millions and millions of people every single day right
508
because that's a very practical way in my seat
509
that i can start using this technology to help people in their day -to -day lives not just healthcare discoveries
510
and everything else that we're also doing
511
but i thought that's quite a powerful demonstration of literally your
512
day -to -day customer seeing actually their just day -to -day life get a little bit easier
513
because of something that you know elon demis
514
and others in this room have helped create yeah exactly the
515
the most immediate thing is just being able to ask um like having a very smart friend
516
that you can ask anything um you know ask how to make something how to solve any problem
517
and it'll tell you um so and obviously companies are going to adopt this
518
so i think maybe you'll have much better customer service, I guess, essentially, that'll probably be the first thing you notice.
519
And then we talked about education.
520
So having a tutor, so if you're trying to understand a subject, like having a phenomenal tutor on any subject,
521
that's really pretty much there already, almost.
522
I mean, we need to, obviously, AI needs to stop hallucinating before, you know, it can't give you...
523
I mean, we still have a little bit of the problem where it can give you an answer that's confidently wrong
524
with great grammar and, you know, bullet points and everything in citations that was not real.
525
So it has to be okay.
526
We need to make sure it's not giving you confidently wrong tutor answers.
527
But that's going to happen pretty quickly where it is actually correct. So, yeah.
528
I was going to say, for any parent who was homeschooling during COVID and realizing what their kids needed to be helped with, that will come as an enormous relief, I think.
529
Very, very good.
530
Right, have we got, let's go questions over here.
531
Who have we got?
532
Have we got any microphones?
533
Brent, are you there?
534
Perfect.
535
Hi, Brent Huberman.
536
So, you know, you've spoken eloquently about abundance and the age of abundance.
537
So it feels like, obviously, with AI, it's everything everywhere, everywhere, all at once.
538
But with robots, and to get the age of abundance, we'll need a lot of robots.
539
I know you're working on robots as well.
540
Are there sort of constraints that we should think of
541
and our politicians should be thinking of
542
that one country might get heavily behind in robots
543
that can do all these things and enter the age of abundance and therefore be at a strategic disadvantage?
544
Well, really anything that can be actuated by a computer is effectively a robot.
545
So you can think of, frankly, Tesla cars are robots on wheels.
546
Anything that's connected to the internet is effectively an endpoint actuator for artificial intelligence.
547
So you've got Boston Dynamics, obviously they've been making impressive robots for a while.
548
I think they're at this point mostly owned by Hyundai,
549
so I guess Hyundai is probably going to make robots that are humanoid and some rather interesting shapes that I wasn't anticipating,
550
like the one that has wheels and looks sort of like a kangaroo on wheels.
551
I'm going to show what that is.
552
Looks a little demanded, frankly.
553
But there's going to be all sorts of robots.
554
You've got the company Dyson in the UK, which I think does some pretty impressive things.
555
I think the UK will not be behind, actually, on that front.
556
UK also has ARM, which is really the best, one of the best, perhaps the best,
557
in chip design in the world.
558
Tesla uses a lot of ARM technology.
559
Almost everyone does, actually.
560
So I think UK is in a strong position.
561
Germany obviously makes a lot of robots, industrial robots.
562
I think generally countries that make robots of any kind, even if they seem somewhat conventional, will be fine.
563
I do think there is a safety concern, especially with humanoid robots, because at least the car can't chase you into this building,
564
not very easily, or chase you off a tree or, you know, you can sort of run up a flight of stairs and get away from a Tesla.
565
I think there's a Stephen King movie about that, if your car gets possessed.
566
So, but if you have a humanoid robot, it can basically chase you anywhere.
567
So I think we should have some kind of hardwired local cutoff that you can't update from the Internet.
568
So anything that can be software updated from the Internet obviously can be overridden
569
but if you have a local sort of off switch where you perhaps say a keyword
570
or something and then
571
that puts the robot into a safe state some kind of
572
localized safe state ability an off switch you know where you
573
don't have to get too close to the robot I don't know
574
so we do if we've got millions of these things going all over the place you're not selling it just you
575
No, I know.
576
I'm saying this is something we should be quite concerned about because if Robert can follow you anywhere,
577
then what if they just one day get a software update and they're not so friendly anymore?
578
Then we've got a James Cameron movie on our hands.
579
It's funny you say that because in our session that we had today, I would say who, they made exactly the same point right then so we're talking about,
580
they're talking about movies actually without mentioning James Cameron they're talking about James Cameron movies and they're saying, if you think about it it's not just those movies, but any of these movies,
581
trains subways, metros cars, buses, they said all these movies with the same plot fundamentally all end with the person turning it off,
582
right, or finding a way to shut the thing down, and they were making the same point that you were about the importance of actual physical off -switches, and so all the technology is great
583
but fundamentally this same movie is played out 50 times we've all watched it
584
and it all fundamentally you know you know the point I'm
585
referring to right it all ends in pretty much the same
586
way with someone finding a way to just yeah do the thing at
587
which is a kind of interesting
588
that you said a similar point right yeah it's not the it's not the obvious place you'd go to
589
but maybe that could be one of the tests for the AI we should say like
590
blank is your favorite James Cameron movie fill in the blank yeah excellent right um yes we got over there yep perfect
591
hi a question for you both um so i'm a founder of a ai
592
and ml scale up in the third center for ai
593
which is leeds in the north of england i'm a bit
594
biased um since the launch of chat gpt three months after
595
that we saw a real increase in phishing attacks using much
596
more sophisticated language patterns um what do we do to protect businesses consumers
597
that they trust this technology better and how do we bring them along that journey with us
598
well i think we shouldn't trust it that much actually um
599
it is actually quite quite a significant challenge
600
because we're getting to the point where even open source ai can pass human capture tests
601
so you know this is are you a human identify all the traffic lights in this picture, you're like, okay.
602
Yeah, it's going to have no problem doing that.
603
In fact, it'll do it better than a human and faster than a human.
604
So we're like, how do you, you know, at the point at which it's better, a better human, better passing human tests than humans, then, well, what tests actually make sense?
605
That is a real problem.
606
I don't actually have a good solution to it.
607
One of the things we're trying to figure out on the X platform is how to deal with
608
that because we really are at the point where even with open source, you know, readily available AI, you don't need to be sort of leading the field.
609
You can actually be better than humans at passing these tests.
610
And that's sort of why we think, well, perhaps we should sort of charge a dollar or a pound a year.
611
It's a very tiny amount of money, but it still makes it privately expensive to make a million bots.
612
So, and especially if you need a million payment methods, then you run out of sort of stolen credit cards pretty quickly.
613
So that's sort of where we're thinking, like, we might have to sort of just charge some very tiny amount of money, 0 .3 cents a day,
614
effectively, to deal with the onslaught of AI -powered bonds.
615
And that is still a growing problem, but it will be, I think, perhaps an insurmountable problem next year.
616
so and then you have to worry about well manipulation of information is making something seem very popular
617
when in fact it is not because it's getting boosted by all these likes
618
and and reposts from ai powered bots
619
so that's why i sort of think somewhat inevitably it leads
620
to some small payment in order to uh dramatically increase the cost of a bot um so i think i think Frankly,
621
I think probably any social media system that doesn't do that will simply be overrun by bonds.
622
I think my general answer would be we need to show
623
that we are on top of mitigating the risks so people can trust the technology.
624
That's what actually the last couple of days has been about
625
on the Safety Summit is just showing we're investing in the Safety Institute, having the people who can do the research on these things to figure out how we mitigate against them,
626
And we have to do it fast and we have to keep iterating it
627
because all of us probably in this room believe that the technology can be incredibly powerful, but we've got to make sure we bring people along that journey with us that we're handling the risks that are there.
628
And as I said, there's a job to do.
629
And the last couple of days, I think we make good progress on it because we want to focus on the positives and manage these things.
630
But that requires action.
631
And that's what the last couple of days have been about
632
your your story your analogy there was part of the research
633
that actually you know the team working on the task force here published
634
and presented yesterday i don't know if you saw it was uh
635
which is a essentially
636
that it was a using ai to do to create a ton of fake profiles on social media
637
and then infiltrate particular groups yeah with particular information yeah and And at the moment, that is to say, to your point, it's like cost -free.
638
It's getting to the point where it's like, really, you're going to have 100 for a penny sort of thing.
639
Ridiculous.
640
And if you think about some of these social networks at quite a neighborhood or town level, it's not that many fake profiles that you need to quickly create.
641
Suddenly they're everywhere, and there's some local issue that might be of importance.
642
And the team have run versions of how that would look like, and suddenly they're interacting with everybody and then spreading misinformation around.
643
Yeah, that's a real challenge.
644
Literally, as part of the research that we published on misinformation yesterday, it's a real challenge.
645
Yeah, exactly, and to your point, I mean, the images, you don't even need to steal somebody's picture, because that's traceable, but you can actually just,
646
say, create a new image of a person, realistic -looking, but doesn't exist, and then create a biography,
647
realistic, but doesn't exist, and do that en masse, and practically the only way to be able to tell is that the grammar is too good.
648
to give away no typhos? come on
649
I'm getting waved at
650
because I think we are out of time I know we take one very brief last question
651
and let's make a good one yes sir, go on, you're right in front of me, go on thank you for the opportunity Elon, question for you
652
related to X platform are there simple things we can do especially
653
when it comes to visual media you alluded to the fact
654
that it's fairly straightforward and effectively free to make people like yourself say and do things that you never said or did.
655
Can we do something like cryptographically signed media?
656
I'm from Adobe.
657
We're working on this project.
658
Twitter was a member.
659
Love to see X come back.
660
Digitally signed media to indicate not only what was created by AI, but what came from a camera, what was real, to imbue a sense of trust in media that can go viral.
661
That sounds like a good idea, actually.
662
Some way of authenticating would be good.
663
so yeah that sounds like a good idea we should probably do it
664
there you go
665
actually on that point so I've already this is particularly pertinent for people in my job
666
and I've already had a situation happen to me with a doctored image that goes everywhere negative,
667
by the time everyone realises well that's fake and we should stop sending it the damage is done
668
and actually we were again reflecting today
669
if you think next year you've got elections in you know
670
i think you know the us india i think indonesia um probably here there you go massive news um
671
and actually you've got just an enormous chunk of the world's population is voting next year right
672
and you've got eu elections as well yeah you know actually
673
just these issues are right in front of us you know
674
next year is where big elections across the globe probably the
675
first set of elections where this has been a real issue yeah
676
so figuring out how we manage
677
that is I think mission critical for the people who want
678
the integrity of our democracy yeah I mean it's quite interesting
679
like the Pope in the puffer jacket have you seen
680
that one that's amazing
681
but I mean I still run into people who think that's
682
real I'm like well what are the odds he's wearing a pop a jacket in July in Rome, you know, be sweating.
683
But it actually looked cool.
684
Quite dashing, I have to say.
685
In fact, I think AI fashion is going to be a real thing.
686
So I've been doing gloom.
687
We live in the most interesting times, and I think this is, it is 80 % likely to be good and 20 % bad,
688
and I think if we're cognizant and careful about the bad part, on balance, actually, it will be the future that we want, or the future that is preferable.
689
And it actually will be somewhat of a leveler, an equalizer, in the sense that, you know, I think everyone will have access to goods and services and education.
690
And so, you know, I think probably it leads to more human happiness.
691
So I guess I'd probably leave on an optimistic note.
692
Perfect.
693
Yeah.
694
Well, that is a great note to end on.
695
I think, you know, we all want that better future.
696
We think it's there.
697
The promise of it is certainly there.
698
Lots of people in this room, including yourselves, are working hard to make it happen.
699
Our job in government is to make sure it happens safely.
700
But on the basis of this conversation in the last couple of days, I'm certainly leaving more confident that we can make that happen.
701
It's been a huge privilege and pleasure to have you here.
702
Thank you very much for having me.
تنزيل التطبيق
كل ما تحتاجه للتحدث بطلاقة
تقييم النطق بالذكاء الاصطناعيتقييم كل جملة
تدريب IPAأتقن كل صوت
المفرداتابنِ رصيدك من الكلمات
لعبة المفرداتتعلّم أثناء اللعب
نفس القناة
FOR YOU
فيديو موصى به
C12
154 جمل
歐巴馬談 AI：喊放慢的公司「是真的在擔心自己的產品」【AI 大人說】
Daily ConversationWorking Adults
B21
394 جمل
The Story of Elon Musk
Daily ConversationWorking Adults
B21
76 جمل
Nvidia's New Laptops - RTX Spark
Daily ConversationWorking Adults
B21
260 جمل
Roadster Event
Daily ConversationWorking Adults
B21
123 جمل
Google Just Dropped NEW AI Updates!🔥
Daily ConversationWorking Adults
B21
257 جمل
Is AI About to Change Movies Forever?
Daily ConversationWorking Adults
B21
25 جمل
Elon was my only customer of our Al Supercomputer #jensenhuang #elonmusk #openai
Business EnglishDaily ConversationWorking Adults
B24
132 جمل
Are we heading to an era of stagnation or innovation?
Daily ConversationUniversity StudentsWorking Adults
B24
182 جمل
BBC News Today: Top Stories You Need to Know 14-09-2026 Breaking News
Daily ConversationUniversity StudentsWorking Adults
B22
174 جمل
One Woman’s Search for Meaning Led Her to AI Delusion
Daily ConversationWorking Adults
C11
40 جمل
I MADE $15B WITH CHAT GPT
Business EnglishDaily ConversationWorking Adults
C11
85 جمل
Significant slowdown in AI payoff could tip economy into a recession, says Apollo Global's Slok
Daily ConversationWorking Adults
المفردات وملاحظات التحدث لهذا الدرس
يحتوي هذا الفيديو على 702 جملة و9093 كلمة للتدرب على الترديد. مدة الكلام 51:09. يتكلم المتحدث بسرعة طبيعية، نحو 178 كلمة في الدقيقة، قريبة من المحادثة اليومية. 86% من الكلمات ضمن أكثر 3,000 كلمة شيوعًا في الإنجليزية، ويُستحسن مراجعة الباقي قبل البدء.
أهم المفردات في هذا الفيديو
15 كلمة من الفيديو تستحق التعلّم، مع النطق والمعنى:
| الكلمة | النطق | المعنى |
|---|---|---|
| robot اسم | /ˈɹoʊ.bɑt/ | رُوبُوت, رُوبُوط |
| abundance اسم | /əˈbʌn.dn̩s/ | كَثْرَة, غَزَارَة |
| innovation اسم | /ˌɪn.əˈveɪ.ʃən/ | اِبْتِكَار |
| bias اسم | /ˈbaɪ.əs/ | تَحَيُّز |
| transparency اسم | /tɹænsˈpæɹənsi/ | شَفَّافِيَّة |
| referee اسم | /ˌɹɛf.əˈɹi/ | حَكَم |
| deploy فعل | /dɪˈplɔɪ/ | عَبَّأَ, نَشَرَ |
| vast صفة | /væst/ | شَاسِع, وَاسِع |
| founder اسم | /ˈfaʊ̯n.dəː/ | مُؤَسِّس, مُؤَسِّسَة |
| chase فعل | /t͡ʃeɪs/ | طَارَدَ |
| compete فعل | /kəmˈpiːt/ | تَنَافَسَ, تَبَارَى |
| regulation اسم | /ˌɹɛɡjʊˈleɪʃən/ | تَنْظِيم |
| reward اسم | /ɹɪˈwɔɹd/ | مُكَافَأَة, ثَوَاب |
| summit اسم | /ˈsʌmɪt/ | قِمَّة |
| artificial صفة | /ˌɑː.tɪˈfɪʃ.əl/ | مُتَصَنَّع, صِنَاعِيّ |
الأفعال المركبة التي ستسمعها
| الكلمة | النطق | المعنى |
|---|---|---|
| come back فعل | عَادَ, رَجَعَ | |
| come out فعل | /ˌkʌm ˈaʊt/ | خَرَجَ, غَادَرَ |
| come up فعل | طَلَعَ | |
| give up فعل | تَنَازَلَ, تَخَلَّى | |
| hands up | /hæn(d)ˈzʌp/ | اِسْتَسْلِمْ |
القواعد في هذا الفيديو
أكثر التراكيب التي يستخدمها المتحدث، مع العبارات نفسها من الفيديو:
| التركيب | في الفيديو |
|---|---|
| المضارع التام المستمر have/has been + -ing — حدث بدأ سابقًا وما زال مستمرًا | you've been doing · have been talking · we've been talking |
| الجمل الشرطية if + جملة، will/would + فعل — شرط ونتيجته | if we've got a skilled population they'll be · if they seem somewhat conventional, will be |
| المضارع التام have/has + اسم المفعول — حدث ماضٍ ما زال له أثر الآن | has done · i've been · I've never really had |
| المبني للمجهول be + اسم المفعول — التركيز على ما يحدث لا على من يفعله | was immersed · being engaged · are released |
نطق يحتاج إلى انتباه
يستخدم المتحدث 208 من الصيغ المختصرة، مثل we're, I'm, you've. انطقها بصيغتها القصيرة كما تسمعها.
- أصوات “th”: mathematics /mæθ(.ə)ˈmæt.ɪks/, algorithm /ˈælɡəɹɪðm̩/
- أصوات “sh” و“zh”: innovation /ˌɪn.əˈveɪ.ʃən/, regulation /ˌɹɛɡjʊˈleɪʃən/, artificial /ˌɑː.tɪˈfɪʃ.əl/, essentially /ɪˈsɛnʃəli/, dish /dɪʃ/
- الكلمات الطويلة — انتبه لموضع النبر: innovation /ˌɪn.əˈveɪ.ʃən/, capability /ˌkeɪ.pəˈbɪl.ɪ.ti/, transparency /tɹænsˈpæɹənsi/, universal /ˌjunəˈvɜɹs(ə)l/, regulation /ˌɹɛɡjʊˈleɪʃən/
كيف تتدرب بهذا الفيديو
- استمع إلى الفيديو كاملًا مرة واحدة دون أن تتكلم، ودوّن الكلمات التي لا تعرفها.
- ابدأ بسرعة 0.75×، وردّد جملة جملة، ثم عُد إلى السرعة العادية عندما يصبح الأمر سهلًا.
- سجّل صوتك وقارنه بالأصل، مع الانتباه إلى كلمات مثل robot, abundance, innovation.
ما هي تقنية التظليل الصوتي؟
التظليل الصوتي (Shadowing) تقنية تعلم لغة مدعومة علمياً، طُورت أصلاً لتدريب المترجمين الفوريين المحترفين. الطريقة بسيطة لكنها قوية: تستمع لصوت إنجليزي أصلي وتكرره فوراً بصوت عالٍ — كظل يتبع المتحدث بتأخير 1-2 ثانية. تُظهر الأبحاث تحسناً كبيراً في دقة النطق والتنغيم والإيقاع وربط الأصوات والاستماع والطلاقة.