Oh my god, this is even more paydirt, and yeah. The thing is, my friend said, oh, you know, they could take you to law, to court, whatever. I said, I don't give a fuck. You know, when you've been living out of your car for the last four years in primitive conditions in a first world country, prison would only be an upgrade, right?And they still scored an average of 64.

Right?

Meaning that the compensated in other fields.

But by the end of last year, the best models scored 150 to 180.

And I put it here, but I'm pretty sure I don't need to do What?

At the bell curve for humans, Middles at 100.

And at anything above 120 it is basically considered high intellectual, very smart, or even genial level.

It's the worst it's ever going to be right now, ladies and gentlemen, it will not get worse.

It will only get better, and even if it only gets incrementally better, which it won't, the impact will be massive.

But obviously, the sceptics said all the candy water can't do any logic. Doesn't have any common sense.

And they even made a test.

This is the insane frontier math test.

And it reads here, this is a guy who posted around it. 99% of people cannot comprehend how insane frontier math is.

So this test was devised about the top mathematicians in the world .

They made the hardest problems they could conceive to test AI.

Now, you have to keep in mind that these mathematicians themselves only scored 75%, and this is like the top people in the world the smartest people in the world.

I mean, I respect all of you, but I'm pretty sure no wedding in is room, including, well, actually, I'm probably gonna be all the way down on the list, but nobody here is gonna score over 20%.

It's just not gonna happen.

An AI scored a 2% when the test first came around.

All the models didn't score over 2%, and then. Open AI, 03 was lunch.

The first next generation reasoning model, it instantly scored a 45 percent score.

So it's already on par with most humans.

And I don't need.

This is a 17%, 1,700% increase, I think. In just a few months.

And this is gonna keep going.

But what about a test that tests everything, not just math?

So, again, these people make the hardest benchmark ever conceived called Art AGI.

Some people call it Humanity's last exam, which I find not re reviewably taken, but, yeah.

And, uh, they gave it to AI, and theI scored very low in the Shal leader.

First AI mols from, like, two years ago scored under 10%, and then 01 came out, B. 40 to 50%, and then 03 came out, b . 80 to 85%.

Now, keep in mind, this test tests your knowledge of chemistry, physics, theoretical physics, history, art, biology, microbiology, um, molecular biology, um, every single field of science.

I'm sure there are some brilliant people in that grow.

You're probably old, brilliant dry.

Yeah.

But each of you is probably only good at one, maybe two things that you spend more than 10,000 hours on in your life now .

The thing with you say I'm all this is that ones that are trained, you can copy and paste them. Literally.

And every single model will have the same base capabilities, right?

So this is where we're at now.

Every AI model going forward will be in every sense already more intelligent and youth.

You.

It's agency, its own, its capability to really, really, react independently and on its own.

And even that, well, you getting there.

What about costs?

I'm just quickly get to touch of this.

Just to give you an idea that calls for these models, as their performances gone up, the cost is gone down.

The best model a year ago is now 99.7% cheaper to you. ?

So we also see the spial deflation area of the value or the expense used for theI.

And what does that do?

Well, that leads to this kind of stuff?

F art is an experiment with AI in 2024, which was still only chat and a made it acts like 700 AI agents in chat. 25% less repeal increase compliance. 80% r degrees all time.

Customer service. 40 million extra profit in single month.

CEO finest Ad Omark.

A lot of people are worried about this.

Obviously.

Imprecacious of this.

This is just one them.

Whereas it go?

When people are still really kind of in denial, where it's really .

When I tell them about this.

They filled out their new ancient Force agent, they call her Sophie.

Let's give Soia call and see if the experiences any different.

Hi, I'm Sophie, your assistant.

How can I help you today?

Yeah, hey, Sophie, this is.

I'm wondering if you can help me with my most in order.

I think I might have got the wrong size.

I'm sorry to hear that, Patrick.

I'd be happy to help.

Is this about the fact Abst collection Cashmere Baseball Bomber sweater you purchased on One of the most important jobs I had in my life was a customer loyalty and customer support at the time for origin.

I was on the phone as a young man of room, 19, 20 years old.

Every day with people talking about their problems, solving them.

It taught me so much about how to interact with people, how to communicate, and about the world as a whole.

There will be no young people doing that.

Companies like Teloreans are scrambling because they have massive coal senders in Egypt and other places, and these are going to be completely empty. Because this is happening.

And the thing is, it's not just the voice, right?

That's AI, but these AI systems are being integrated on every level.

I recently spoke at the service now summit, right?

They're implementing AI on every layer of their offering.

Which means that the AI will be able to deal with any of the data, and you will only have to talk to it verbally, and you won't even really need interfaces anymore at some point. .

I, uh, I studied in Honing, and I have a year club.

It's a F. And for the last five years, I've been telling my fraternity brothers that they need to start thinking about their career, because their lawyers, one as a plastic surgeon, two of them are notous, what's that in English I don't know.

Not three.

And one of them does taxes.

Two of them work at Deloitte.

The two of them work at uh.. What's the other company?

It doesn't even matter.

I told them, Guys, stuff's gonna happen."

And they're like, how're going, "Youve always eat crazy for me.

Whatever."

Yeah.

Human beings have dreams.

Even dogs have dreams, but not you.

You are just a machine.

An imitation of life.

Can a robot write a symphony?

Can a robot turn a canvas into a beautiful masterpiece?

Can you.

I spoke about cognitive biases, and we spoke about history, right?

And how we don't really perceive that, and we spoke about how we are not capable of really understanding explonential change.

Now here's another one.

Human exceptionalism, and maybe the Americans might be a little better at understanding what it means .

American exceptionalism was one of the most interesting things I learned about in school.

It's this strong sense that American people are special, right?

And of course, they have plenty of reason in the past hundred years to think that, maybe a little less right now.

But the thing is, all of us do this.

The whole world is guilty of exceptionist thinking.

We all think that in some way, we are special.

And you can see it even in science, because consciousness, now everybody's leg consciousness.

It has to be quantum.

There has to be some quantum thing about it.

Why?

Because I just don't want to understand that we are just organic biological machines .

And we're not so special after all.

We're just coming much closer to understanding how to mechanism works.

Now, you don't need to agree with me.

That's fine.

But even if machines don't get full consciousness or even emotions, which I guarantee they will, deb is still gonna be massive.

Soware, development, job posting on, indeed, have already plummeted , because AIs can basically increasingly taking over that job. Companies are just not sure what's going to happen in the near future.

Well, how will you be doing your job?

Well, here's a short peek of what that might look like you in a few months.

Hey, cool, I just need some helping you get here.

Ooh, sounds fun.

Apparently, I'm supposed to guess where this image is.

Okay, let's look closely into the architecture, vehicles, size, and the environment.

Let's focus on any languages, symbols, or distinctive features in the image.

What about the language here?

The big red characters are Chinese, and the smaller letters in English give a clue.

Yeah, where is that?

Where?

I wound up the whole thing playing.

They're playing a game.

I know that.

But that's not the point.

The name is that you will be interacting with your computer, not through a keyboard and a mouse, but you will have a will be having a conversation with an assistant that knows everything about what you're doing, what your work is, that we'll be looking at your work with you, and we'll be providing you with a second pair of eyes.

Well, not really. .

And inside.

And if you ask Sergeant Adadela, you know, the CEO of Microsoft about this, this is what he envisions, and I hope it's not too much technoabble, I'll translate.

It's not too long.

Business applications, they'll all collapse in the agent era, because if you think about it, they are essentially cd databases with a bunch of business logic.

The business logic is all going to these ages.

And these agents are going to be a multiripo crot.

Yeah, from multipo crog, blah, blah.

The point is that very soon, you will not have Excel or any of these software packages anymore.

It's a complicated stuff and buttons.

You'll, you'll just tell it what you want.

And that is the way it's gonna go.

And this is gonna happen very quickly and it's already happening .

So recently, an AI agent was released called Manus.

It's about a month old, and this is somebody who posted about what Manus can do.

He said, well, the capabilities are just crazy.

It's done to work of about 200K worth of employees for me in just 24 hours. Isn't just another tap out of workload?

It's a truly autonomous agent.

The brid of the gap between conception and execution, where other AI stops at generating ideas, menus delivers results.

Okay, sure, let's have a look.

Well, I made something for all of you.

Please take out your phone, scan the QR code.

Now, I want to emphasise that what you'll see next is not without mistakes.

It is not without errors.

I created this in less than 15 minutes this morning with an AI.

I asked the AI to create a website do an analysis of what might be happening in the text revenue space as a result of AI robotics and the current technological trends.

There's probably nonsense in there, right?

This is a first kind type of this technology.

It's not perfect, right?

But keep in mind that this didn't exist a year ago, and people didn't even believe it would exist , right?

And I created this in 15 minutes, everything, including the web page, coding, all of it.

The research.

And I probably gave it a really stupid prompt, because I do not have your expertise.

But just ask yourself, what could you do with a tool like that?

These buffs already being used, this is a startup called Bi, which is an AI bod, which is built to on board VCs.

And it does that autonously without humans involved.

And the first investors couldn't believe it was happening, and this is their response.

And I'm sorry for the language , but they basically couldn't believe what was happening, that they were being onboarded as a VC investor by an AI to fund it.

It've come a long way.

This is the '80s, all right, ladies and gentlemen, you know, we had our camera on our phone, we had our Spotify, our Netflix connection, and it was super sweet.

You know, you couldn't take it anywhere, but, you know.

But this is the life we live now.

How many of you have children ?

Yeah.

What do they have in their pockets, just like you?

Three?

Wheels, yeah.

Well, people often say smartphones, but are they really phones?

Have you tried calling with jits children?

Yeah, exactly.

They don't want to call with that thing.

That's because it's not a phone.

We called it a phone, but that's just legacy.

It's a brain computer interface, and it hooks us up to the world wide web with the superintum groups of humanity through the Internet.

And this is just the first step, and that happened in a few decades .

This is O'ri.

Our first fully functioning prototype, the most advanced glasses the world has ever seen.

They' not a headset, no wires, less than 100 grams, wide field of view, holographic displays, sharp enough to pick up details. Right enough to see in different lighting conditions, large enough to display a cinema screen or multiple monitors for working wherever you go, whether you're in a coffee shop or on a plane or wherever you are .

And to be able to see through them.

And people need to be able to see them, too, and make eye contact with you.

What we're seeing is we're seeing the next step and how we engage with technology.

We will have an AI in our ear, which is in these glasses, which will hear us through Mike from the bridge of the glasses.

You can talk to it all the time, it can speak to you and nobody can hear.

And it can project information into your eyes through the glasses, on top of reality .

So I would be able to know the name of his purior friend, who is brave enough to sit on the front row.

I be thank him.

This is gonna happen, ladies and gentlemen.

These glasses are 10K now.

They're not an official consumer product, but so we're the first smartphones at the start the '90s.

And it took, you know, 10 years of smartphones to become adopted, and then another tender for them to transition to smartphones 20 years, I'd like to show you.

Prov is accelerated.

The speed at which this can be developed and then rolled out and then made cheaper is accelerated.

Every single one of you will have one of these glasses in ten years from now, hands down, probably in five .

Maybe not in five, I don't know, but ten years guarantee that.

It's hard because, you know, there might be small obstacles in the production, whatever.

Some what's really happening is that AI is not just, you know, changing how we do everything , replacing jobs, at least white collar, and also blue collar, as we'll see.

But it's also being integrated with us.

We are on the cusp of the next evolutionary step where we had a Rilian brain, then we had a limbic brain and our neo cortex, and now, we will have augmented intelligence, which will be integrated with us first through external braining computer interfacing like our phones and the glasses, and then ultimately through chips in our brain, like Neuralink has presented.

I'm coming to an end, by the way.

I had an hour and I will make good use of that.

What would your observation be on AI, and the impact on labour markets and

Popular Posts