在手机上保存，获得更好体验
本字幕由TME AI技术生成
If you haven't stayed up three nights being anxious about ai
If you haven't an extential crisis about it
You probably haven't really experienced ai
It is a weird thing
What's it mean to be human
What's it mean to think what would i do for a living
What will my kids do
What does it mean that is better than me at some of this stuff is this real or is an illusion
Nobody action knows where ai is heading right now and how good it's going to get
But we shouldn't feel like we don't have control over how ai is used as managers and leaders
You get to make these choices about how to deploy these systems to increase human flourishing as individuals
We get to decide how to be the human
Who uses these systems
Well
Ai is here to stay
That is something
That you get to make a decision about how you want to handle and to learn to work with
And learn to thrive with
Rather than to just be scared of i meet themi like a professor
The word in school of the university pennsylvania
Where i studied innovation entrepreneurr ship and artiicial intelligence and the author book coo intelligence
Living at working with a eye artificial intelligence about prediction
Basically ai is a very fancy auto complete for a long time
That was about numerical prediction and doing so of complex algorithms of math
So that netflix could recommend a show for you to watch
Amazon could figure where where decide this next warehouse or testha fifiure ure how to use data to make sure
Its cars were driving automatically the thing
That these systems were bad of predicting was the next word in the sentence
So if your sentence ended with word filed it don't know wheyou're filing your taxes of filing your nails
What happened that was different
Was the innovation of the large language model in two thousand and seventeen a breakthrough paper call the tensions all you need outlined a new kind of ai call the transform with attention mechanism that basically let the ai pay attention it to not just the final word in the sentence
But the entire context
The sentence
The paragraph
The page and so on large language models work by taking huge amounts of information like all the data of the internet
There a lot of hariry potter fan fiction for example
Because that's what the internet contains and based on all of us data
The ai go through
Through process called pretraining
And this is that really expensive part that only a few companies the world can do
And during that time the air learns relationships between words or parts of words called tokens
So it learns that key
We and strawberry cloclosely related
But theok and potato are not closely related
Ted they learns across thousands dimensions that that llthat dimenmension space
We can understand that it didn't do predictions tions it turns out
Ununexpectively
When large language models get big enough
They also do all kinds of other things
We didn't expect it
Thedidn't expect in regood in medicine
But they're actually quite good and be doctors and many circumstances that it didn't expect in regood at creativity
But they can generate ideas better than most humans can
And so they're general purpose models
They do many different things aa interesting ly gpt doesn't just stand for the gpt and chat gpt
It also stand for general purpose technology
Which is one of these once in your generation technologies things like steam power or the internet or electrivi
Or that how everything they touched the alter society they alter how we work
How they alter how we relate to each other in ways that are really hard to anticipate
So you can't think it certainties
You should think it's scenarioand there's really four scenenos in the future
The first is actually i think the least likely which is that the world is sttic
That that is is the best airor ever going to use
I think that's unlikely in fact
Whatever air or using now is the worst ai or ever going to use
So if the core large language mluent development stop
Right now there's another ten years of just making it work better with tools and with industry in ways that contincontinue be disruptive
So i think that's the dangerous view
Because it isn't static
It's evolving
So i want to skip actually to the last scenario before covering scenarios to indurace the scenario for is agi artificial general intelligence
This is the idea that a machine will be smarter than a human almost all tasks
And this is the explicit goal of open ai
They want to build agi
And there's a lot of debate about what this means when we have a machine smarter than a human
And it can do all human jobs s can create ai smarter than itself
And i think artificial super intelligence asi and humans become expooustely overnight
And there's people genuinely worried about this
And i think it's worth spending little time time being lilily worworry too
Because other people
People But
But i think that
That scenario tends to take agency away from us
Because it's something that happens us
And i think that it's more important to worry about what i call scenariottwo and three
Which is continued linear exponential growth
嗯嗯
We don't know how good ai is going to get right now the doubling time for capabilities about every five to nine months
Which is an exceptly fast doubling time more res law
Which is the rule that's kind of keep the computer world going doubles the power of computer prosting chips every twenty four to twenty six months
So this is a very fast ate of growth
It's very likely that ai will continue improve and get better in the near term
And now is a good time for you to start to forget
How to use ai to support what makes you human or good at things
And what things as ai gets better
Did you might want to start handing off more to the ai
We have a lot of early evidence
That this is going to be a big deal for work
So there's now a multiple studies across fields ranging from consulting to legal to a marketing to programming
Suggesting twenty to eighty percent performance improvements across a wide range of tasks for people who use iyeors is don't the proof ting being human is that were stuck in our heads
And a lot lot of decisions that are bad result from us not having enough perspectives
Ai is very good and cheap way of providing additional perspectives
You don't delicate to with device
But getting a advice
Forcing it reflect for a moment
Forcing very a thgity to reject or acceptive
And i can give you the license to actually be really creative
And don't deraciy your own innovation
So you can ask it to create crazy suggetions tions
You
What's the most complicated way to solve this problem
What's the most expensive way to solve this problem
What's the most worst idea
I have to be this was a superbilant make this problem worse
It can be very innerable to realize at the ai is quite very creativity
And think think us the very human tivbut
But think think we embrace effect that ai can help us be more creative
That actually is very exciting a lot of us feel stiteful to creativity
And having a partner who can work with you and doesn't ever judge
You can often feel very exberating 今天one of the really weird things about large language models s they don't work
Like we think computer should work lm ms are very bad at math
Computer should be good at math
Who large language models are weirdly emotional and you can threaten you or want to be your friend seemingly
And so it can be very hard to know in advance what they're good or about it in fact
Nobody actually knows the answer
We call this the jacket frontier of ai
The idea that there's almost a spiky shape to what the aacan can do
What what can can do
So it can what you need to do is understand the shape of the frontier
You need to know
When the a is likely to lie you
And when it's not going to hallucination of first to the idea that what the ai produces could be entirely made up plausible sounding fake information
The thing about ai is though everything it doesn't hallucation ation there's no mind there you might start to come more persuaded biit
You might become blind to its biases
You might think it's more cable than it is ai kind of works a luted like a psychoc
It's
It's really good at telling you what you want to hear the fact
It's accurate
So often it's kind of weird actually and hallucinrates tes to be dropping over time
And what you need do a sharper in your own intuition with working with a tool
Get a sense of when you see something that might make you concerned the a when you ask people about a future vi
There's a term use within ai insiders called pedoom
Which is a probability that were all going to die
I do not have a pedoom that i really think about
Because i don't think we going to side a probability to things going wrong
And again that makes the technology
The agent we get to decide how this thing is used
And if we think about this the right way
This freezes from boredom and tedium and disaster
But i think we need to think about the mistakes we made in regularating other technologies
You know and what the vangers are what we did rently for the internet versus
Social medil and are decisions
We get to make that are both personal about how we use it at a organizal level
About how it's deployed and at a societal level
And it's not an inevitability
The technology just does what it does
It does what it does
Because society lets it do that get smarter
Faster
With videos from the world's biggest thinkers to learn even more from the world's biggest thinkers get big think plus for your business
