Latest / Elon Musk Thinking / Elon Musk Talk About AI Tech in the Future.

Transcript

Elon Musk Talk About AI Tech in the Future. · Elon Musk Thinking

0:01
Okay, all right, well, good evening, everybody. Welcome Elon,
0:05
thanks for being here, Thank you for having me. We feel very prov
0:08
we're excited to have you. Right, So I'm going to start with some
0:11
questions and then we're going to open it up. Let me get straight into
0:13
it. So Bill Gates said, there is no one in our time who
0:18
has done more to push the bounds of science innovation than you. What's kind
0:21
of going to say, Well, that's a nice thing to have anyone say
0:24
about you. Nice coming from Bill Gates. But oddly enough, when it
0:28
comes to AI, actually for around a decade you've almost been doing the opposite
0:33
and saying, hang on, you need to think about what we're doing and
0:36
what we're pushing here, and what do we do to make this safe?
0:38
And actually maybe we shouldn't be pushing as faster or as hard as we are.
0:42
I mean, you've been doing it for a decade, Like what was
0:44
it that caused you to think about it that way? And you know,
0:47
why do we need to be worried? Yeah? I've been somewhat of a
0:50
Cassandra for quite a while where people would I would tell you, like we
0:55
should really be con learned about aid, like, what are you talking about?
0:57
Like, I've never really had any experience with AI, but since I
1:00
was immersed in technology, I have been immersed in technology for a long time,
1:04
I could see it coming. So but I think this year was that
1:11
there have been a number of breakthroughs. I mean, you know, the
1:14
point of which someone can see a dynamically created video of themselves, you know,
1:19
like it sort you can make a video of you saying anything in real
1:23
time or me, And so there's sort of the deep bake videos which are
1:27
really incredibly good, in fact, sometimes more convincing than real ones and a
1:34
bit deep real and and then and then obviously things like chat GPT we were
1:42
quite remarkable. Now I saw GPT one, GPT two, GPT three,
1:46
GPT four that you know, the whole sort of lead up to that,
1:49
So it was easy for me to kind of see where it's going. If
1:53
you just sort of extrapolate the points on a curve and as seeing that train
1:56
will continue, then we will have profound, artificially intelligence and obviously at a
2:00
level that far exceeds human intelligence. So I'm glad to see at this point
2:07
that people are taking safety seriously and I'd like I say thank you for holding
2:14
this AI safety conference. I think actually will go down in history as being
2:17
very important. I think it's really quite profound, and I do think overall
2:25
that the potential is there for artificial intelligence AI too have most likely a positive
2:32
effect and to create a future of abundance where there is no scarcity of goods
2:39
and services. But it is somewhat of the magic genie problem where if you
2:46
have a magic genie that can grant all the wishes, usually those stories don't
2:51
end well. Who care for what you wish for? Including wishes? Yeah,
2:57
so you talked a little bit about the summit and thank you for being
3:02
engaged in it, which has been great and people enjoyed having you there.
3:06
Puspending in this dialogue now, one of the things that we achieved today in
3:09
the meetings between the companies and the leaders was an agreement that externally, ideally,
3:16
governments should be doing safety testing of models before they're released. I think
3:22
this is something that you've spoken about a little bit. It was something we
3:24
worked really hard on because you know, my job in government is to say,
3:29
hang on, there is a potential risk here, not a not a
3:30
definite risk, but a potential risk of something that could be bad. My
3:36
job is to protect the country and we can only do that if we develop
3:39
the capability we need in our safety Institute and then go in and make sure
3:43
we can test the models before they are released. Delight of that that happened
3:46
today. But you know, what's your view on what we should be doing,
3:50
right you've talked about the potential risk? Right again, we don't know,
3:53
but you know, what are the types of things governments like our should
3:55
be doing to manage and mitigate against those risks. Well, generally think that
4:00
it is good for government to play a role when the public safety is at
4:04
risk. So you know, really, for the vast majority of software,
4:10
the public safety is not at risk. I mean, if if the app
4:13
crashes on your phone, your laptop, it's not a massive catastrophe. But
4:17
when you're talking about digital superintelligence, I think, which does pose a risk
4:24
to the public, then there is a role for government to play to safeguard
4:28
the interest of the public. And this is of course true in many fields,
4:31
you know, aviation, cars, you know, I deal with regulators
4:36
throughout the world because of stalling being communications, rockets being aerospace, and cars
4:44
you know, being tracked vehicle transport. So I'm very familiar with dealing with
4:46
with regulators, and I actually agree with the vast majority of regulations. There's
4:50
a few that I disagree with from time to time, but zero point one
4:53
percent, probably less than one percent of regulations I disagree with. So there
4:59
is some concern and from people in Silicon Valley who who've never dealt with regulators
5:03
before, and they think that this is gonna just crush innovation and slow them
5:09
down and be annoying. But and and it will be annoying. It's true,
5:15
they're not wrong about that. But but I think there's we've learnt over
5:19
the years that having a referee is a good thing. And if you look
5:26
at any sports game, there's always a referee, and and nobody's suggesting I
5:30
think to have a sports game with that one. And I think that's the
5:34
right way to think about this, is for for governed to be a referee,
5:39
to make sure the sportsmanlike conduct and that the public safety is you know,
5:47
it is addressed that we care about the public safety. Because I think
5:50
there might be at times too much optimism about technology. And I speak I
5:55
say that as a technologist. I mean, so I ought to know and
6:00
and like said, on balance, I think that that the AI will be
6:04
a forcible good most likely, but the probability of it going bad it is
6:10
not zero. Said, so we just need to mitigate the downside potential.
6:14
And then how you talk about referee and that's what we're demonstrate. Yeah,
6:17
well there we go. I mean, you know, and we talked about
6:19
this and Demos and I discussed this a long time ago, like literally facing
6:23
right and actually Demister his credit and the credit of people in the industry,
6:30
did say that to us. I think you say, it's not right that
6:32
Demis and his colleagues are marking their own homework, right, there needs to
6:34
be someone independent, and that's why we've developed the Safety Institute here. I
6:40
mean, do you think governments can develop the expertise. One of the things
6:43
we need to do is say, hang on, you know, Demis and
6:46
Sam all the others have got a lot of very smart people doing this.
6:48
Governments need to quickly tool up capability wise, personnel wise, which is what
6:54
we're doing. I mean, do you think it is possible for governments to
6:56
do that fast enough, given how quickly the technology is developing, or what
7:00
do we need to do to make sure we do do it quick enough.
7:02
No, I think it's good. It's a great point you're making. The
7:05
pace of AI is faster than any technology I've seen in history by far,
7:13
and it seems to be growing in capability by at least fivefold, perhaps tenfold
7:17
per year. It'll certainly grow by an automatitude next year. Yeah, so
7:26
so, and government isn't used to moving at that speed. But I think
7:30
even if there are not firm regulations, even if there's not even if there
7:35
isn't an enforcement capability, so we're having insight and being able to highlight concerns
7:40
to the public will be very powerful. So even if that's all that's accomplished,
7:46
I think that will be very very good. Okay, yeah, well
7:48
hopefully we can do better than that, hopefully. Yeah. Yeah, No,
7:51
but that's how fast we were talking before. It was striking. You're
7:55
someone who spent their life in technology. They're living More's law. And what
8:00
it was interesting over the last couple of days talking to everyone who's doing the
8:03
development of this, and I think you concur with this is is just the
8:07
pace of advancement here is unlike anything all of you have seen in your careers
8:13
and technology. Is that fair because you've got these kind of compounding effects from
8:16
the hardware and the data and the personnel. Yeah, I mean the two
8:26
currently the two leading centers for a development are the San Fransco Bay area and
8:30
the sort of London area, And there are many other places where it's being
8:33
done, but those are two leading areas. So I think if you know,
8:37
if the United States and the UK and China are sort of aligned on
8:45
safety, that's all going to be a good thing. That's really that's where
8:48
that's where the leadership is generally. You actually you mentioned China there. So
8:52
I took a decision to invite China to summit over the last three days,
8:56
and it was not an easy decision. A lot of people criticizes me for,
9:01
you know, my view as if you're going to try serious conversation,
9:03
you need to. But what would your thoughts? You do business all around
9:05
the world. You just talked about it there. Should we be engaging with
9:09
them? Can we trust them? Is that the right thing to have done?
9:13
If we don't, if China is not on board with AI safety,
9:16
it's somewhat of a moot situation. The single biggest objection that I get to
9:22
any kind of AI regulation or sort of safety controls are well, China is
9:26
not going to do it, and therefore they will just jump into the lead
9:30
and exceed us all But actually China is willing to participate in AI safety and
9:37
thank you for inviting them, and they you know, I think we should
9:43
thank China for attending. When I was when I was in China earlier this
9:48
year, my main subject of discussion with this the leadership in China was a
9:52
safety and saying that this this is really something that they they should care about,
9:58
and they took it seriously and and and you are too, which is
10:01
which is great. And having them here I think was essential. Really if
10:05
the if they're if they're not participants, it's it's pointless. Yeah, that's
10:11
and I think we were pleased. I think they were engaged yesterday in the
10:13
discussions and actually ended up signing the same communicy that everyone else did, which
10:18
is a good stop, right. And I said, if we need everyone
10:20
to approach us in a similar way, if we're going to have a realistic
10:24
chance of resolving it. I was going to tell you you talked about innovation
10:26
earlier and regulation being annoying. There was a good debate today we had about
10:31
open source, and I think you've kind of been a proponent of algorithmic transparency
10:37
and making some of the ex algorithms public and you actually we were talking about
10:41
every hint and on the way in. You know, he's particularly been very
10:46
concerned about open source models being used by bad actors. You've got a group
10:50
of people who say they are critical to innovation happening in that distributed way.
10:54
Look, it's it's a trick. There's probably no perfect answer, and there
10:58
is a tricky balance your thoughts on how we should approach this open source question
11:03
or where should we be targeting whatever regulatory or monitoring that we're going to do
11:09
well. The open source algorithms and data tend to lag the closed source by
11:16
six to twelve months. So that but given the rate of improvement that there's
11:22
actually therefore quite a big difference between the closed source and the and the open
11:26
If things are improving by a factor of let's say five or more than being
11:31
a year behind is you're five times worse. So it's a pretty big difference.
11:35
And that might be actually an okay situation, but it certainly we'll get
11:41
to the point where you've got open source AI that can do that, that
11:46
will start to approach human level intelligence, will PHAPS succeeded. I don't know
11:52
quite what to do about it. I think it's somewhat inevitable. There will
11:54
be some amount of open source, and I guess I would have a slight
11:58
bias towards open source because at least you can see what's going on. It
12:03
wears closed source. You don't know what's going on. Now, it should
12:05
be said with AI that even if it's an open source, do you actually
12:09
know what's going on? Because if you've got a gigantic data file and you
12:13
know sort of billions of data points or weights and parameters, you can't just
12:22
read it and see what it's going to do. It's a gigantic file of
12:26
inscrutable numbers. You can test it when you when you run it, you
12:30
can test it. You can run a bunch of tests to see what it's
12:33
going to do. But it's probabilistic as opposed to deterministic. It's not it's
12:39
not like traditional programming, where you've got it you've got very discrete logic and
12:45
the outcome is very predictable, and you can read each line and see what
12:48
each line is going to do. A neural net is just a whole bunch
12:54
of probabilities. I mean, it sort of ends up being a giant Comma
12:58
separated value file. It's like our digital guide is a CSP file. Really
13:05
okay, but that's kind of what it is. Yeah, Now, that
13:11
point you've just made is one that we have been talking about a lot,
13:13
because again conversations with the people are developing their technology make the point that you've
13:18
just made. It is not like normal software where there's predictability about inputs improving
13:24
leading to this particular output improving, and as the models iterate and improve,
13:28
we don't quite know what's going to come out the other end. I think
13:31
Demis would agree with that, which is why I think there is this bias
13:35
for Look, we need to get in there while the training runs are being
13:37
done, before the models are released, to understand what is this new iteration
13:41
broad about in terms of capability, which it sounds like you would would agree
13:46
with. I was going to shift Goz a little bit on. You know,
13:48
you've talked a lot about human consciousness human agency, which actually might strike
13:52
people as as strange given that you are known for being such a brilliant innovator
13:58
and technologist, but it's it's quite heartfelt when I hear you talk about it
14:01
and the importance of maintaining that agency in technology and preserving human consciousness. Now
14:07
it kind of links to the thing I was going to ask, is when
14:09
I do interviews or talk to people out and about in this job about AI,
14:13
the thing that comes up most actually is it probably not so much the
14:16
stuff we've been talking about, but jobs. It's what is AI mean for
14:20
my job? Is it going to mean that I don't have a job or
14:24
my kids are not going to have a job. Now, you know,
14:26
my answer as a policymaker, as a leader is actually, AI is already
14:31
creating jobs, and you can see that in the companies that are starting.
14:35
Also, the way it's being used is a little bit more as a co
14:39
pilot necessarily, versus replacing the person. There's still human agency, but it's
14:43
helping you do your job better, which is a good thing. And as
14:46
we've seen with technological revolutions in the past, clearly there's change in the labor
14:52
market the amount of jobs. I was quoting an MIT study today that they
14:56
did a couple of years ago is something like sixty percent of the jobs at
14:58
that moment exists forty years ago, so hard to predict. And my job
15:01
is to create an incredible education system, whether it's at school, whether it's
15:05
retraining people at any point in their career, because ultimately, if we've got
15:09
a skill population, they'll be able to keep up with the pace of change
15:13
and have a good life. But you know that it's still a concern.
15:16
And what would your kind of observation be on AI and the impact on labor
15:22
markets and people's jobs and how they should feel about that as they think about
15:26
this. Well, I think we are seeing the most disruptive force in history
15:31
here. You know, we have for the first time, we will have
15:35
for the first time something that is smarter than the smartest human and that I
15:41
mean, it's hard to say exactly what that moment is, but there will
15:46
come a point where no job is needed. You can have a job if
15:50
you want to have a job for sort of personal satisfaction, but the AI
15:56
will be able to do everything. So I don't know if that makes people
16:00
comfortable uncomfortable, it's it's you know, That's why that's why I say,
16:06
if you, if you, if you wish for a magic genie that gives
16:10
you any wishes you want, and there's no limit. You don't have those
16:11
three limits three wish limits, not since you just have many as many wishes
16:15
as you want. So it's both good and bad. One of the challenges
16:22
in the future will be how do we find meaning in life if if you
16:27
have a magic genie that can do everything you want. I do think we's
16:32
it's hard. You know, when there's when when there's new technology, it
16:33
tends to have usually follow an S curve. In this case, we're going
16:38
to be on the exponential portion of the S curve for a long time,
16:44
and we have to like so we able to ask for anything. And it
16:47
won't be we won't have universal basic income. We'll have universal high income.
16:52
So in some in some sense it will be somewhat of a leveler or an
16:55
equalizer, because really I think everyone will have access to this magic Jenny anyway,
17:02
able to ask any question. It will certainly be good for education,
17:04
you could it'll be the best tutor you could add them the most patient tutor,
17:08
UH said there all day. And there will be no shortage of goods
17:15
and services, will be an age of abundance. I think if i'd recommend
17:21
people read in Banks, the Banks culture books are probably the best in visioning.
17:27
In fact, not probably, they're definitely by far the best envisioning of
17:30
an a future. There's nothing even close. So I'd recommend, really recommend
17:36
Banks. I'm a very big fan. All these books are good. There's
17:41
not Jay which one all of them. So so that's that'll give you a
17:47
sense of what is a I guess a fairly utopian or pro topin future with
17:56
with Ai, which is good for as you said, it's a universal high
18:02
income, which is a nice phrase, and that's it's good from a kind
18:03
of materialistic sense of abundance. Actually, that then leads to the question that
18:08
you pose, right, I'm someone who believes, you know, work gives
18:11
you meaning. I write a lot about that. As you know, I
18:14
think work is a good thing. It, you know, gives people purpose
18:17
in their lives. And if you then remove a large chunk of that,
18:21
you know, what does that mean? And where do you get that?
18:22
You know, where do you get that drive, that motivation, that purpose?
18:26
I mean you're talking about it. You work a lot of hours,
18:29
you know, as I was mentioning, when we were talking earlier. I
18:32
have to somewhat engage in liberate suspension of disbelief because I'm putting so much blood,
18:37
sweat and tears into a work project and burning the you know, three
18:40
am oil. Then I'm like, wait, why am I doing this?
18:45
I can just wait for the AI to do it. I'm just lashing myself
18:48
for no reason. Must be a gutton for punishment or something. So we
18:56
called called called demos and tell them to hurry up and then you can have
18:59
a holiday. It's the plan. Yeah, No, it's a Look,
19:03
it's a tricky. It's a tricky thing because I think you know, part
19:04
of our job is to make sure that we can navigate to that very I
19:10
think largely positive place that you're described people through it between now and then,
19:15
because these things bring a lot aboutter change in in the labor market as we've
19:18
seen. Yeah, I think it's probably is generally a good thing because you
19:22
know, there are a lot of jobs that are uncomfortable or dangerous or which
19:26
sort of tedious, and the computer will have no problem doing that. We're
19:30
happy to do that all day long. So you know, it's it's fun
19:34
to cook food, but it's not that fun to wash the dishes, but
19:37
the computer is perfectly happy to watch the dishes. I guess there is.
19:42
You know, we still have sports like where where humans compete and like the
19:48
Olympics, and obviously a machine can can go faster than any human, but
19:52
we still have we still humans race against each other and and have a you
20:00
know, these sports competitions against each other where even though the machines are better,
20:03
which I guess computnitity, who can be the best human at something?
20:07
And people do find performing in that. So I guess that's perhaps a good
20:11
example of how even when machines are faster than are stronger than us, we
20:14
still find a way we saw enjoy computing against other humans, so at leaste,
20:18
who was the best human? Yeah, that's a good it's a good
20:21
analogy. And we've been talking a lot about managing the risks. Just before
20:25
we move on and finish on AI is just talk a little bit about the
20:29
opportunities. You know, you're engaged in lots of different companies. You're an
20:33
obvious line which is doing which is doing some exciting stuff. You touched on
20:38
the thing that I'm probably most excited about, which is an education. Yeah,
20:42
and I think many people will have seen sal CON's video from earlier this
20:47
year. Is Ted talk about as you talked about, it's like a personal
20:48
tutor. Yeah, an amazing personal tutor, an amazing personal tutor. And
20:52
we know the difference in learning having that personal tutor is incredible compared to class
20:57
from learning. So if you can have every child, I'll have a personal
21:00
tutor specifically for them that then just evolves with them over time. Yeah,
21:03
it could be extraordinary. And so that you know, for me, I
21:07
look at it, I think, gosh, that is within reach at this
21:08
point. And that's one of the benefits I'm most excited about. When you
21:12
look at the landscape of things that you see as possible, what is it
21:15
that you know you are particularly excited about. I think certainly AI AI tutors
21:22
are going to be amazing, perhaps already are. I think there's also perhaps
21:27
companionship, which may seem odd because how can the computer really be your friend?
21:32
But if you if you have an AI that has memory, you know,
21:36
and remembers all of your interactions and has read everything, you've got to
21:40
actually like give it permission to read everything you've ever done, so it really
21:41
will know you better than anyone, perhaps even yourself, and and and where
21:47
you can talk to it every day and those conversations spoiled upon each other,
21:51
you will actually have a great friend as long as that friend can stay your
21:56
friend and not get turned off or something. Don't turn off my friends.
22:03
But I think that will actually be a real thing. And I have a
22:08
one of my sons is sort of has some learning disabilities and has trouble making
22:12
friends actually, and I was like, well, you know he AI friend
22:17
would actually be a great fri enid Okay, you know's that was a surprising
22:21
answer. That's actually it's worth worth reflecting on that. That's really interesting.
22:25
I mean we're already seeing it actually as we deliver you know, psychotherapy anyway,
22:30
now doing far more by digitally and by telephone to people, and it's
22:33
making a huge difference. And you can see a world and which actually,
22:37
you know, AI can provide that social benefit to people. Just a quick
22:41
question on on X and then we should open it up to everybody. You
22:45
made a change when in one of the made many changes, but one of
22:51
the one of the changes, We've got a real thing about it. You
22:56
really do, really do one of the changes which you know kind of you
23:00
know, it goes into the space that you know, we have to operate
23:03
in. And this, this balance between free speech and moderation is you know,
23:07
we grapple with as politicians. You were grappling with your own version of
23:11
that, and you you moved away from a kind of manual human way of
23:15
doing it, the moderation to the community notes and yeah, and I think
23:21
that's you know, it was an interesting change, right, It's not what
23:22
everyone else has done. It would be good, you know, what's what
23:26
was the reasoning behind that? And why do you think that is a better
23:29
way to do that? Yeah? Part of the problem is if you if
23:33
you empower people as sensors, then well there's going to be some amount of
23:37
bias they have, and then whoever appoints the sensors is effectively in control of
23:42
information. So then the idea behind community notice, well, how do we
23:47
have a consensus driven I mean, so it's not really censoring it, but
23:52
consensus driven approach to truth? How do we how do we how do we
23:56
make things the least amount untrue? Like you say, like what you can't
24:03
perhaps get to pure truth, but you can aspire to be more truthful So
24:08
the thing about community notes is it doesn't actually delete anything. It simply adds
24:12
context. And that context could be this thing is untrue for the following reasons.
24:18
And but but importantly with community notes, everything is open source actually,
24:23
so you can see the software, every line of the software. You can
24:27
see all of the data that went into a community note, and and you
24:32
can independently create that community note. So if you've got if you see a
24:36
manipulations of data, you can actually highlight that and say well this, this,
24:40
this, the appears to be some gaming of the system, and you
24:42
can suggest improvement. So it's it's it's maximum transparency, which is i think
24:48
combined with the kind of wisdom of the crowds and transparency to get to a
24:52
better answer is you. And really one of the key elements of community notes
24:56
is that in order for a note to be shown, people who have historic
25:00
disagreed must agree. And there is a bit of AI usage here, so
25:04
there's will populate a parameter space around each contributor to community notes, and then
25:12
the parameter space so everyone's got basically these these vectors associated with them, which
25:18
so it's it's not as simple as as right or left it's saying it's it's
25:22
more, it's several hundred vectors. That that because things are more complicated than
25:26
something right right or left and and and then will we'll do sort of inverse
25:33
correlation, say like, okay, these these people generally disagree, but they
25:37
agree about this note. Okay, so then that so then that that that
25:41
gives the note credibility. Yeah, that's that's the that's the core of it.
25:45
And it's working quite well. I get to see a note actually be
25:49
be present for more than a few hours that that is incorrect. So the
25:53
batting average is extremely good. And when I ask PEP, people say,
25:56
oh, they're worried about community notes sort of being dis like, send me
26:00
one and then they can't. So so I think it's I think it's quite
26:04
good. I mean, the general aspiration is with the X platform is to
26:10
inform and entertain the public and to be as accurate as possible and as truthful
26:14
as possible, even if someone doesn't like the truth. You know, people
26:18
don't always like the truth, not always, but that's that's that's the aspiration.
26:26
And I think if we are if we stay true to the truth,
26:30
then I think we'll find that people use the system to learn what is going
26:33
on and to learn. I think actually truth pays, so I think it'll
26:41
be what I mean, assuming you don't want to engage your self delusion,
26:45
then I think it's it's the smart move. So excellent, very helpful.
26:52
Right, let's open it up to all our guests here, and we've got
26:55
some microphones that they'll come. Put your hands up, they'll come and find
26:57
you. We got yes, go for it. Thank you, good evening.
27:03
Alice Bentink from Entrepreneur. First, thank you for a fascinating conversation.
27:07
I suppose a question for each of you, Prime Minister. The UK has
27:11
some of the best universities in the world, we have the talent. What
27:15
will it takes the UK to be a real breeding ground for unicorn companies?
27:21
And elon being a founder in the UK is still a non obvious career choice
27:26
for the most exceptional technical talent. What are the cultural elements that we need
27:30
to put into place to change this? Thank you both. Well you want
27:33
to go first, go for it? Sure? Well, you're right that
27:37
there are cultural elements where you know, the culture should celebrate creating new companies
27:45
and there should be a bias towards supporting small, small companies because they're the
27:53
ones that need nurturing. The larger companies really don't need nurturing. So you
28:00
know, just you can think of it's sort of like a garden. If
28:02
it's a little sprout that needs needs nurturing, if it's a mighty oak,
28:04
it doesn't need quite as much. So I think that that is a mindset
28:10
change that is important. But but I should mention that London is you know,
28:18
London and San Francisco or the Bay Area are really the two census for
28:21
AI. So that so London is actually doing very well on that front.
28:26
That the two I said, the two leading locations on Earth. You know,
28:30
San Francisco's probably head of London, but London is really very strong.
28:33
But London Area, greater London home counties, I guess, keep keep going.
28:42
So I'm just saying objectively this is the case. And but you do
28:48
need that. You need you need the infrastructure, need, you need landlords
28:52
who are willing to rant to new companies, who you need law firms and
28:56
accounts that are willing to support new companies. And it's generally it is a
29:00
mindset change, and I think some of that is happening. But I think
29:04
really it's just culturally people need to decide this is this is a good thing.
29:08
Yeah yeah, actually, well thanks for what you said about the UK.
29:11
It's something that we work hard on. Lots of people in the room
29:15
are part of what makes this a fabulous place for companies, including Alice.
29:19
So that's what i'd say is my job is to get all the you know,
29:25
the nuts and bolts right, make sure that all of you are starting
29:26
companies can raise a capital that you need everything from you know, your seed
29:30
funding with our incredible you know eis tax release all the way through to you
29:36
know, your late stage rounds. And we need reform of our pension funds
29:38
and the Chancellor's got a bunch of incredible reforms to unlock capital from all the
29:44
people who have it and deploy it into growth equity. Right, that is
29:47
a work in progress. We're not there yet, but I think we're we're
29:49
making We're making good progress. We need talent, we need people, you
29:52
know, so that means an education system that prioritizes the things that matter.
29:56
And you've seen my reforms. I go on about more matt, more maths,
30:00
more maths, but I think that it's important, but also attracting the
30:03
best and the brightest here. If you look at our fastest growing companies in
30:07
this country, and I think it's probably the same in the US, over
30:11
half of them have a non British founder, right, And so that tells
30:15
you we've got to be a place that is open to the world's best and
30:18
brightest entrepreneurial talent. So the visa regime that we've put in place, I
30:22
think, does that makes it easy for those people to come here. And
30:25
then actually it's the thing that we've spent the beginning of the session talking about
30:29
the regulation, right, making sure that we've got a regulatory system that's pro
30:33
innovation. That Yeah, of course we always need guardrails on the things that
30:37
will worry us, but we've got to create a space for people to innovate
30:41
and do different things. Now, those are all my jobs. The thing
30:44
that is tougher is the thing that Elon talked about, which is culture.
30:47
Right, It's how do you transpose that culture from places like Silicon Valley across
30:52
the world where people are unafraid to give up the security of a regular paycheck
30:57
to go and start something, be comfortable with failure, You talk about that
31:02
a lot. I think you talk about it more and when you're playing games,
31:03
right, but that you've got to be comfortable failing and knowing that that's
31:07
just part of the process. And that is as a tricky cultural thing to
31:11
do overnight, but it's an important part of I think creating that kind of
31:15
environment. Yeah, if you don't succeed with your first startup, it shouldn't
31:18
be a sort of a catastrophic career ending exactly, thing it should be you
31:22
know, well good I think generally should like should be like, well,
31:25
you know, you gave it a good shot, you know, and and
31:26
and now try again and exactly. It's so one thing I'm going to mention
31:33
is like, obviously creating a company is sort of a high risk, high
31:34
reward situation. But I don't know quite what the how it works in the
31:41
UK, I think probably better than continental Europe, but the stock options are
31:48
very difficult in most parts of Europe. I'm not sure how does you give
31:52
it. But if somebody's basically going to risk the sort of life savings and
31:56
with and the vast majority of startups fail, So I mean you hear about
32:00
the startups that succeed, but most companies are most startups consist of you know,
32:07
a massive amount of work followed by failure. That's actually most most companies.
32:13
And so it's a high risk kigh reward and so the higher reward part
32:16
does need to be there for it to make sense. I think that was
32:20
a very soft pitch for a tax policy that I need chance that. But
32:23
actually I can tell you so like AI agree, and we have so we
32:28
have, I think relative to certainly European countries, but certainly the US,
32:31
definitely California a much lower rate of capital gains tax right. So for those
32:37
people who are risking and growing something like, we think the reward should be
32:39
there at the end. So it's twenty percent capital gains tax right. And
32:44
on stock options, I don't know if we've got anyone from Index Ventures in
32:47
the room. So Index one of our bleeding VC funds here. They do
32:52
a regular report looking at most countries' tax treatment of stock options. And you
33:00
know, when I was a Chancellor of Treasury Sexuary equivalent, you know,
33:04
we were I think down at we were pretty good, but we were fourth
33:07
or fifth. Then I said we need to for exactly the reason that you
33:08
mentioned was like this has got to be the best place for renovative We need
33:12
to move that up. And I think in the last iteration of that report
33:15
we had because of the changes that Jeremy and I have made, we have
33:19
moved up to I think second from memory. Hopefully that actually give you and
33:22
everyone else some comfort that we recognize. That's important because when people work hard
33:27
and risk things, yeah, they should be able to enjoy the rewards high
33:30
reward. Yeah. And I think we have a we very much have a
33:32
tax system that supports that. And those are the values that you know,
33:36
I believe in, and I think most of us in this room probably do
33:38
as well. Right, next, next question, I've got seven front of
33:42
me, and then I'll come over here. Gone gone, So thanks very
33:47
much. We've talked about some really big ideas, global changing ideas. I'm
33:53
really interested, particularly in the Contexte creation of science and technology, superhubs and
33:58
so on. How does that app onto the everyday lives of people living in
34:02
say Austin, Texas, to choose around you and more in my case Nottingham,
34:07
East Midlands. Uh, what what is? How do you see that
34:10
evolving for people? You know every day sort of every day effective AI for
34:19
context, Elon so SEB. SEB runs our equivalent of CDs, right or
34:23
Walgreens, so you know when actually I visited, right, So he's got
34:25
millions of people coming in the shops every day, and it's making sure how
34:30
do we make this relevant? I think is your question, how is this
34:32
relevant to that person? You know? Maybe actually let me go I'll go
34:36
first on that because I think it's a fair a fair point. I was
34:40
just going over with the team a couple of things that we're doing because I
34:43
was saying, like, how are we doing AI right now that it's making
34:45
a difference to people's lives. And we have this thing called gov dot UK,
34:50
which is which actually when we when it happened several years ago, was
34:53
a pioneering thing. All the government information brought together on one website, gov
34:58
dot uk, so you need to get a driving license, passport, any
35:01
interaction with government. It was centralized in a very easy, relatively easy to
35:06
use way, better than most. So we're about so we're about to deploy
35:14
AI across that PI form. So that is something that I think, you
35:16
know, several million people a day use right, so a large chunk of
35:21
the population is interacting with gov dot UK every single day to do all these
35:24
day to day tasks. Right, every one of your customers is doing all
35:28
those things, and so we're about to deploy AI into that to make that
35:32
whole process so much easier. Because you know, some people will be like,
35:36
look, well I'm currently here and I've lost my passport and my flights
35:39
in five hours. You know at the moment that would require you know how
35:44
many steps to figure out what you do? You know, well, actually,
35:45
when we deploy the AI, it should be that you could just literally
35:49
say that and boom boom boom, bom boom. This is what we're going
35:52
to do walking through it, and that's going to benefit millions of millions of
35:54
people every single day, right, Because that's a very practical way in my
35:59
seat, that I can start using this technology to help people in their day
36:01
to day lives, not just healthcare discoveries and everything else that we're also doing.
36:07
But I thought that's quite a powerful demonstration of literally your day to day
36:10
customer seeing actually their just day to day life get a little bit easier because
36:15
of something that you know, Elon Demis and others in this room have helped
36:16
create. Yeah, exactly, The most immediate thing is just being able to
36:21
ask, like having a very smart friend that you can ask anything, you
36:27
know, ask how to make something, how to solve any problem, and
36:30
it'll tell you so. And obviously companies are going to adopt this, so
36:37
I think maybe you'll have much better customer service. I guess essentially that'll probably
36:40
be the first thing you notice. And then if we talked about education,
36:50
so having a tutor, so if you're trying to understand a subject, like
36:55
having a phenomenal tutor on any subject, is that that's really pretty much there
37:01
already almost. I mean we need to obviously, I need to stop hallucinating
37:05
before you know it can't give you. I mean that we still have a
37:09
little bit of the problem where it can give you an answer that's competently wrong
37:14
with great grammar and you know, in bullet points and everything in citations,
37:19
it was not real. So that's be okay. We need to make sure
37:22
it's it's not it's not it's not giving you competently wrong tunor answers. But
37:28
but that's going to happen pretty quickly where it is actually correct. So yeah,
37:34
I'm gonna say for any for any parent who was homeschooling during COVID and
37:37
realizing what their kids needed to be helped with. That will come as an
37:42
enormous relief. I think, yeah, very very good. Right have we
37:45
got let's go questions over here? Who have we got? We only microphones
37:49
or brand you they're perfect? Hi, Brent Herman. So you know you've
37:53
spoken eloquently about abundance in the Age of abundance, So it feels obviously with
37:59
AI it's everything else were everywhere all at once. But with robots, and
38:05
to get the age of abundance, we'll need a lot of robots. I
38:07
know you're working on robots as well. Are there sort of constraints that we
38:12
should think of our politician should be thinking of that we might get one country
38:15
might get heavily behind in robots. They can do all these things and end
38:19
to the age of bunds and therefore be it a strategic disadvantage. Well,
38:25
really, anything that can be actuated by a computer is effectively a robot.
38:30
So you can think of frankly, Tesla cars or robots on wheels. Anything
38:37
that's connected to the Internet is effectively an endpoint actuator for artificial intelligence. So
38:47
you've got Bust and Dynamics. Obviously, they've been making impressive robots for a
38:51
while. I think there are at this point mostly on by Hundai, So
38:55
I guess who is probably gonna make robots of that are humanoid and and some
39:01
interesting shapes that I wasn't just patting, like the one that looks like a
39:06
has wheels and looks sort of like a kangaroo on wheels. I'm sure what
39:08
that is, but it looks a little demanded, frankly. But but there's
39:16
gonna be all sorts of all sorts of robots. You've got the company Dison
39:22
in the UK, which I think there's some pretty impressive things. I think
39:29
the UK will not be behind actually on that front. UK also has ARM,
39:32
which is really the best one one of the best paths, the best
39:39
in chip design in the world. Tees I use a lot of a lot
39:45
of ARM technology. Almost everyone does, actually, so I think UK is
39:51
in a strong position. Germany obviously makes a lot of robots, natural robots
39:57
that I mean, I think generally countries that make robots of any kind,
40:01
even if they seem somewhat conventional'll be We'll be fine. I do think there
40:08
is a there is a safety concern, especially with humanoid robots, because at
40:15
least the car can't chase you into this building not very easily, you know,
40:19
or chase you over a tree, or you know, you can sort
40:22
of run up a flight of stairs and get away from a tesla. I
40:28
think it's a Stephen King movie about that if your car gets possessed. So,
40:34
but if you have a humanoid robot, it can it can basically chase
40:37
you anywhere. So I think we should have some kind of hardwired local cutoff
40:45
that you can't update from the internet. So anything that can be software updated
40:50
from the internet obviously can be overridden. But if you have a local sort
40:53
of off switch where you have to say a keyword or something, and then
40:57
that puts the robot into a safe state, some kind of localized safe state
41:04
ability, an off switch, you know, where you don't have to get
41:09
too close to the robot. I don't know, if we've got millions of
41:13
these things going over the place, you're not selling it, just you know,
41:17
I know, I'm saying this is something we should be quite concerned about
41:22
because Robert Robert can follow you anywhere. Then you know what if they just
41:28
one day get a software update, and they're not so friendly anymore. We've
41:32
got a James Cameron movie on our house. It's actually that's it's funny you're
41:37
saying that because in our session that we had today, you know, just
41:42
I would say who they made exactly the same point right then, as we're
41:45
talking about they're talking about movies actually about mentioning James Cameron. They're talking about
41:50
James cameeron movies, and they're saying, if you think about it's not just
41:53
those movies, but any of these movies, trains, subways, metro.
42:00
They said, all these movies with the same plot fundamentally all end with the
42:04
person turning it off right or finding a way to shut the thing down.
42:08
And they were making the same point that you were about the importance of actual
42:13
physical off switches and so all the technology is great, but undamentally, the
42:16
same movie has played out fifty times, we've all watched it, and all
42:20
fundamentally, you know, you know, the point I'm revenged right or ends
42:23
in pretty much the same way with someone finding a way to just do the
42:27
wrong. Which is kind of interesting that you've said a similar point. Right,
42:30
it's not the it's not the obvious place you'd go to, but be
42:34
One of the tests for the AI was just so like, blank is your
42:37
favorite Dreamers camera movie? Excellent? Right, Yes, we got over there.
42:47
Yeah, perfect, Hi question for you both. So I'm a founder
42:53
of a AI and mL scale up in the third center for AI, which
42:58
is leads in the North of England a bit biased. Since the launch of
43:01
chat GBT three months after that, we saw a real increase in phishing attacks
43:07
using much more sophisticated language patterns. What do we do to protect businesses consumers
43:13
that they trust this technology better and how do we bring them along that journey
43:16
with us? Well, I think we shouldn't trust it that much. Actually,
43:25
it is actually quite quite a significant challenge because we're getting to the point
43:29
where even open source AI can pass human capture tests. So you know this,
43:34
are you a human identify all the traffic lights in this picture? You're
43:37
like, Okay, it's going to have no problem doing that. In fact,
43:40
it'll do it better than the human and faster than human. So we're
43:45
like, how do you you know at the point which is better a better
43:47
human? Better passing human tests than humans? Then well, what tests actually
43:52
make sense? That is a real problem. I don't actually have a good
43:55
solution to it. That's one of the things we're trying to figure out on
43:59
the XP form is how to deal with that, because it really we really
44:02
are at the point where even with with open source, you know, readily
44:08
available AI, you don't need to be sort of leading in the field.
44:12
You can actually be better than humans at passing these tests. And that's sort
44:16
of why we think, well, perhaps we should sort of charge a dollar
44:20
or a pound a year. It's a very tiny amount of money, but
44:22
it's it's still it still makes it privatively expensive to make a million bots.
44:28
So and especially if you need a million payment methods, then you run out
44:31
of sort of stolen credit cards pretty quickly. So that that's that's sort of
44:39
where we're thinking, like we might have to sort of just charge some very
44:42
tiny amount of moneyo point three cents a day effectively to deal with the the
44:49
onsought of AI powered bonds. And if if and that that is that is
44:54
spilit growing problem, but it will be I think perhaps an insurmountable problem next
44:59
year. So and then you have to worry about, well, manipulation of
45:05
information is making something seem very popular when in fact it it's not because it's
45:08
getting boosted by all these likes and reposts from AI powered barts. So that's
45:15
why I sort of think somewhat inevitably it leads to some small payment in order
45:21
to dramatically increase the cost of a bart. So I think I frankly,
45:28
I think probably any social media system it doesn't do that. Also believe you
45:30
over on my bart. You know, I think my general answer would be,
45:37
you know, we need to show that we are on top of mitigating
45:40
the risks right so people can trust the technology. That's what actually the last
45:45
couple of days has been about on the Safety Summit is just showing, you
45:49
know, we're investing in the Safety Institute, having the people who can do
45:52
the research on these things to figure out how we mitigate against them, and
45:55
we have to do it fast, and we have to keep iterating it because
46:00
all of us probably in this room, believe that the technology can be incredibly
46:02
powerful, but we've got to make sure we bring people along that journey with
46:06
us, that we're handling the risks that are there and as there's a job
46:09
to do, and the last couple of days I think we make good progress
46:13
on it. Because we want to focus on the positives and manage these things.
46:16
But that requires action, and that's what the last couple of days has
46:20
been about your your story, you know, analogy. There was part of
46:24
the research that actually, you know, the team working on the task force
46:29
here published and presented yesterday. I don't know if you saw it was,
46:34
which is essentially that it was a using AI to do to create a ton
46:39
of fake profiles on social media and then infiltrate particular groups with particular information.
46:46
And actually, at the moment that, as I said to your point,
46:49
there's a cost free it's getting. It's getting at the point where it's like,
46:52
really, you're going to have one hundred for a penny sort of thing.
46:54
Ridiculous. And if you think about some of these social networks that quite
46:58
a neighborhood or town levels, not that many big profiles that you need to
47:01
quickly create. Suddenly they're everywhere and there's some local issue that might be of
47:06
importance, and you know, the team have run versions of how that would
47:09
look like, and suddenly they're interacting with everybody and then spreading and misinformation around.
47:15
Were literally as part of the research that we published on this information yesterday.
47:19
It's a it's a real challenge. Yeah, exactly to your point.
47:22
I mean the images. You don't even need to steal somebody's picture because then
47:27
that's traceable. But you can actually just say, create a new image of
47:30
a person realistic looking but doesn't exist, and then create a biography realistic but
47:37
doesn't exist, and do that en mass and practically the only way of Bill
47:40
tells that the grammar is too good. Don't give away typho. Come on
47:49
now, I'm getting waved at because I think we are out of time.
47:51
I don't we take one very brief last question and let's make it go on?
47:54
Yes, sir, going, you're right in front of me, go
48:00
elon. Question for you related to the X platform. Are there simple things
48:05
we can do, especially when it comes to visual media. You alluded to
48:07
the fact that it's fairly straightforward and effectively free to make people like yourselves say
48:13
and do things that you never said or did. Can we do something like
48:15
cryptographically signed media. I'm from Adobe, We're working on this project. Twitter
48:20
was a member. Love to see X come back. Okay, digitally signed
48:23
media to indicate not only what was created by AI, but what came from
48:28
a camera, what was real to imbue a sense of trust in media that
48:30
can go viral. That sounds like a good idea. Actually, so if
48:36
some way of authenticating would be would be good. So, yeah, that
48:43
sounds like a good idea, which we'll probably do it. Yeah, there
48:46
you go, actually on that on that point. So I've already and this
48:52
is particularly pertinent for people in my job, right, and I've already had
48:55
a situation happened to me with adopted image that goes everywhere negative. By the
49:00
time everyone realizes, well that's fake and we should stop sending it, the
49:04
damages the damage is done. And actually we were again reflecting today. If
49:08
you think next year, you've got elections in you know, I think you
49:13
know, the US, India, I think Indonesia probably here there you go,
49:20
massive news. Actually you've got just an enormous junk of the world's population
49:25
is voting next year, right, and you've got EU elections as well.
49:30
You know. Actually, just these issues are right in front of ours.
49:32
Next year is one of big elections across the globe, probably the first set
49:37
of elections where this has been a real issue. So figuring out how we
49:42
manage that is I think kind of mission critical for the people who want,
49:45
you know, the integrity of our democracy. Yeah, I mean some of
49:47
it it's quite entertaining, Like the pope and the puffa jacket. Have you
49:51
seen that one? That's amazing, But I mean I saw ritter people who
49:55
think that's real. I'm like, well, one of the answer is wearing
49:59
a puffa jacket and lie, you know, be sweating. But it actually
50:04
looks quite quite dashing of things. In fact, I think AI fashion is
50:08
going to be a real thing. So do in glue like, we live
50:12
in the most interesting times, and I think this is it is, you
50:15
know, eight like eighty percent likely to be good and twenty percent bad.
50:17
I think if we're cognizant and careful about the bad part, on balance,
50:22
actually it will be the future that we want. Well for the future that
50:27
is preferable, and it actually will be some somewhat of a leveler, an
50:31
equalizer in the sense that you know, every I think everyone will have access
50:36
to goods and services and education, and so you know, I think probably
50:39
it leads to more human happiness. So I guess I'd probably leave on an
50:44
optimistic note. Tough act. Yeah, that's well, that's that is a
50:49
great note to end on. I think that you know we're all onant that
50:52
better future. I think it's there. The promise of it is certainly there.
50:54
Lots of people in this room, including yourselves, are working hard to
50:58
make it happen. Our job in government is to make sure happened safely.
51:00
But on the basis of this conversation in the last couple of days, I'm
51:04
certainly leaving more confident that we can make that happen. It's been a huge
51:07
privilege and pleasure to have you here. Thank you very much for having me.
51:13
I think it's