We've been told time and again that the

massive large language models trained by

companies like OpenAI and Anthropic are

poised to utterly transform our world.

We've been told that huge percentages of

existing jobs are soon to be automated.

We've been told that skills like writing

and photography and film making are all

about to be outsourced. And we'd be told

if we're not careful, the systems built

on these models might someday soon

become sentient and even threaten the

existence of the human race. But here's

the thing. One of the AI pioneers who

helped usher in this current age is not

convinced. His name is Yan Lakun and

he's been long arguing that not only

will LLM based AI fail to deliver all of

these disruptions, but that it is, and

I'm quoting him here, a technological

dead end.

People have started to listen. Earlier

this month, a syndicate of investors,

including Jeff Bezos and Mark Cuban,

along with a bunch of different VC

firms, raised over a billion dollars to

fund Lacun's new startup, Advanced

Machine Intelligence Labs, which seeks

to build an alternative path to true AI,

one that avoids LLMs all together.

After all of the hype and stressed and

handrigging around LLMbased tools like

CHAP, GPT, and Claude Code, is it

possible that Yan Lun was right that

those specific types of tools won't

change everything? And if so, what's

going to come next? If you've been

following AI news recently, you've

probably been asking these questions.

And today, we're going to seek some

measured answers.

I'm Cal Newport and this is the AI

reality check.

Okay, so here's the plan. I've broken

down this discussion into three sub

questions. Sub question number one, what

exactly is Yan Lun up to and how does

this differ from what the existing major

AI companies are doing? Sub question

number two. How is it possible that he

could be right about LLM running out of

steam if everything we've been hearing

recently from tech CEOs and news media

is about how fast LLMs are advancing and

how this techn is about to change

everything? And number three,

if Lun is right, what should we expect

to happen in the next few years? And

what should we hap expect to happen in

the maybe decade time span? All right,

so that's our game plan here. It's going

to get a little technical. I'm going to

put on my computer science hat, but I'll

try to keep things simple, which really

is the worst of both worlds because it

means that the technical people will say

I'm oversimplifying and the nontechnical

people will say I still don't make

sense. So, I'm going to do my best here

to walk this high wire act. Let's get

started with our first sub question.

What is Yan Lun up to? All right. Well,

let's just start with the basics. Um, I

want to read a couple quotes here from a

recent article that Cade Mets wrote for

the New York Times discussing what just

happened with Lacun's new company. All

right. So, I'm quoting here. Lacun's

startup Advanced Machine Intelligence

Labs or AMI Labs has raised over $1

billion in seed funding from investors

in the United States, Europe, and Asia.

Although AMI Labs is only a month old

and employs only 12 people, this funding

round values the company at $3.5

billion.

Dr. Lun who's 65 was one of the three

pioneering researchers who received the

Turing Award often called the Nobel

Prize in computing for their work on the

technology that is now the foundation of

modern AI. Dr. Lun has long argued that

LLMs are not a path to truly intelligent

machines. The problem with LLMs, he

said, is that they do not plan ahead.

Trained solely on digital data, they do

not have a way of understanding the

complexities of the real world. quote,

"If you try to take robots into open

environments, into households, or into

the street, they will not be useful with

current technology." End quote. Uh, Mr.

Le Brun, who's the CEO of AMI Labs, told

the Times, "We want to help them reach

new situation, react to new situations

with more common sense." All right, so

that's kind of a a highle summary of

what's going on. Let's get in the weeds

here to really get into the technical

details of what Lacun is saying and how

it differs, how his vision differs from

what the major existing frontier AI

companies are actually doing. All right,

let's start with a basic idea here. If

you're an AI company, you're trying to

build artificial intelligence-based

systems that help people do useful

things. This could be like by asking

them questions with a chatbot or having

the system help you produce computer

code if we're talking about coding

agents. At the core of all these

products needs to be some sort of what

we can call digital brain, something

that encapsulates the core of the

artificial intelligence that your tool

or system is leveraging.

So the major AI companies like OpenAI

and Anthropic have a different strategy

for creating those underlying digital

brains than Yan Lun's new company has.

All right. So what are the existing AI

companies doing? They're all in on the

idea that the digital brain behind these

AI products should be a large language

model. Now, we've talked about this

before. You've heard this before, so

I'll go quick, but it's worth

reiterating.

A large language model is an AI system

that takes this input text and it

outputs a prediction of what word or

part of a word should follow. So if we

want to be sort of anthropomorphic here,

what it's trying to do is that it

assumes the text it has as input is a

real pre-existing text and that what

it's trying to do is correctly guess

what followed that text in the actual

real existing pre-existing text. That's

really what a language model does.

So if you call it a bunch of times, so

you give it input, you get a word or

part of the word as output. You then

append that to your input and now put

the slightly longer input into the

language model, you get another word or

part of a word. And if you add that to

the input and put that through the

model, you slowly expand the input into

a longer answer. This is called auto

reggressive text production that you

keep taking the output and putting it

back into the input until the model

finally says, uh, I'm done. And then you

have your your response. So we can think

about it. Then if we zoom out a little

bit, the large language model takes text

as input and then expands whatever story

you told it to try to finish it in a way

that it feels is reasonable.

Under the hood, they look something like

this. Jesse, can we bring this up on the

screen here? Um, this is like a typical

architecture for a large language model.

You have input like here it says to

cats. You that gets broken into tokens.

Those get embedded into some sort of

mathematical semantic space. Don't worry

about that. They then go through a bunch

of transformer layers. Uh each layer has

two sub layers, an intention sub layer

and a feed forward neural network. And

out of the end of those layers comes

some information that goes into an

output head that selects what word or

part of a word to output next. So that's

the it's kind of this linear structure

uh is the architecture of a large

language model. So the way you train a

large language model is you give it lots

of real existing text and what you do is

you knock words out of that text. you

have it try to predict the missing word

and then you correct it to try to uh

make it a little bit more accurate. If

you do this long enough on a big enough

network with enough words, this process

which is called pre-training produces

language models that are really good at

predicting missing words. And to get

really good at predicting missing words,

they end up encoding into those uh feed

feed forward neural network layers

within their architecture lots of

knowledge about the world sort of uh how

things work, different types of tones.

They get really good pattern

recognizers, really good rules. You

actually sort of implicitly

emergently and implicitly within the

feed uh forward neural networks in the

language models, a lot of sort of smarts

and knowledge begins to emerge.

That's the basic idea with a large

language model. So the large the AI

companies their their bet is if these

things are large enough and we train

them long enough uh and then we do

enough sort of fine-tuning afterwards

with post-training

you can use a single massive large

language model as the digital brain for

many many different applications. Right?

So when you're talking with a chatbot,

it's referring it's referencing a the

same large language model that your

coding agent might also be talking to to

help figure out what computer code to

produce. It'll be the same large

language model that your openclaw

personal assistant agent is also

accessing. So it's all about one how

9000 style massive model, massive large

language model that is so smart you can

use it as a digital brain for anything

that people might want to do in the

economic sphere. That is the model of

companies like OpenAI and Anthropic.

All right. So what is Yan Lun's AMI Labs

doing differently? Well, he doesn't

believe in this idea that having a

single large model that implicitly

learns how to do everything makes sense.

He thinks that's going to hit a uh dead

end. That's an incredibly inefficient

way uh to try to build intelligence. And

the intelligence you get is going to be

brittle because it's all implicit and

emergent. you're going to get

hallucinations or sort of odd flights of

uh responses that really doesn't make

sense in the real world. So what is his

alternative approach? Well, he says

instead of having just one large single

model, he wants to shift to what we

could call a modular architecture where

your digital brain has lots of different

modules in it that each specialize in

different things that they're all wired

together.

Let me show you what this might look

like. I'm going to bring on the screen

here a key paper that lacun published in

2022 called a path towards autonomous

machine intelligence. This has most of

the ideas that are behind AMI labs. Um

this paper has this diagram here I have

on the screen. Uh it's an example of a

modular architecture. So he imagines an

AI digital brain now has multiple

modules including a world model which is

separate from an actor which is separate

from the critic which is separate from a

perception module which is separate from

short-term memory which is separate from

an overall configurator that helps move

information between each of these

different modules. So you might have for

example the perception module makes

sense of input it's getting maybe

through text or through cameras if it's

a robot. It passes that to an actor

which is going to propose like here's

what we should do next. But then the

critic is going to analyze it different

options using the world model which has

a model of how the relevant world works

to try to figure out which of these

options is best pulling from short-term

memory. Then the actor can choose the

best of those options which then gets

executed. So it's a much more of a we

have different pieces that do different

things. Now another piece of the yam

lacun image is that you can train

different modules within modular

architecture differently. Again, in a

language model, there's like one way you

train the whole model and all the

intelligence implicitly emerges. In

Lacun's architecture, he says, "Well,

wait a second. Train each module with

the best way uh with whatever way makes

the most sense for what that module

does." So, like the perception module,

let's say it's making sense of the world

through cameras. Well, there we want to

use a sort of uh vision network that's

trained with sort of like classic deep

learning vision recognition of the type

that you know Lun actually helped

pioneer back in the '9s and early 2000s.

But then the world model which is trying

to build an understanding of how the

world works, he's like oh we would train

that very differently. In fact, he has a

particular technique. So if you've heard

of JEA GEA, joint embedding predictive

architecture,

this is a new training technique that

Lun came up with for training a world

model where at a very high level he says

here's the right way to do that. Don't

train a model that tries to understand

how a c a particular domain works. Don't

just train it with the low-level data

like the actual raw words from a book or

raw images from a camera. What you want

to do is take these real world

experiences and convert them all to

high-level representations and train

them on the highle representation. So

like I'm simplifying here a lot. Let's

say you have as input a picture of a

baseball about to hit a window and then

a subsequent picture where the window is

broken. You don't want to train a world

model he argues just on those pictures.

Like if I see a picture like this, the

picture that would follow is one where

the glass is broken. That's how maybe

something like a a a standard LLM style

generative picture generator might work.

He's like instead take both pictures and

have a highle representation. So it's

like a mathematical encoding of like a

baseball is getting near a window. Like

what actually matters? What are the key

factors of this picture and then the

next picture is the window breaks. And

what you really want to teach the model

is when it has this highle setup, a

baseball's about to hit the window. It

learns that leads to the window

breaking. So it's not stuck in

particular inputs but learning causal

rules about how the relevant domain

works. And anyways there's a lot of

other ideas like this the critic and

actor that comes out of RL reinforcement

learning worlds um as sort of well

known. You've you've trained one network

with rewards and another one to propose

actions. And so there's a a lot of

different ideas coming together here.

The third piece about Lacun's vision

that differs from the big AI companies

is he doesn't believe in having just one

system that you train once and is then

the digital brain for all the different

types of things you should do. He says

this architecture is the right

architecture for everything. But you

train different systems for different

domains. So if I want a digital brain

that we can build computer programming

agent tools on, I'm going to take one of

my systems with its world model and

perception and actor and critics and I'm

going to train it specifically for the

domain of producing computer programs

and then all my computer programming

agents that people are building will use

that particular system. But if I want to

do uh help with call centers or

whatever, I might completely train a

different version of the system just to

be really good at call centers. So we

don't have just one massive HAL 9000

that everything uses which is the OpenAI

plan or the anthropic plan. We custom

train systems that maybe all use the

same general architecture but we train

them from scratch for different types of

domains. You're going to get much better

performance out of it. All right. So

that is uh Yan Lun's vision and he says

this is how you're going to get uh much

more reliable and smart and useful

activity out of AI. this idea that we're

just going to train like a massive model

that can do everything based off of just

text. He's like, "Come on, this makes no

sense. That can't possibly be the best,

most efficient route towards actually

having smarter AI." All right, so that

is the key tension between the existing

AI companies and Yan Lun's idea. This

brings us to our second sub question.

How is it possible that Lacun

could be right that LLMs are a dead end

if we've been hearing nonstop in recent

months about how these LLM based

companies are about to destroy the

economy and change everything? How could

we be so wrong?

Lun is not surprised by that. I think

there if we asked him, I'll simulate

Lacun. If we asked him, he would say the

short answer to that question is, look,

a lot of coverage of LLMs recently have

been a mixture of hype and confusing the

specific LLM strategies of the frontier

companies with the idea and

possibilities of AI more generally and

kind of mixing those things together,

which is fine if you're Sam Alman or or

Dario Amade, that's great for you

because you need investment, but it's

probably not the most accurate way to

think about it. Now, if we ask Lacun in

this hypothetical to give a longer

explanation about how we could be so

wrong about LLMs, he would probably say,

"Okay, let me let me explain to you the

trajectory of the LLM technology in

three stages." And I think this will

clarify a lot. All right. So, the first

stage was the pre-training scaling

stage. And this is the stage where the

the AI companies kept increasing the

size of the LLM. So how big those layers

are inside of them, the size of the

LLMs, the amount of data they trained

them on, and how long they trained them.

And there was a period starting in 2020

and lasting until 2024

where making the model bigger and

training them longer demonstrabably and

unambiguously increased their

capabilities. This petered out after

about GPT4. after about GPT4, OpenAI

um we have evidence that XAI had the

same issue. We have evidence that Meta

had the same issue. When they continued

to make their models bigger, they

stopped getting those big performance

jumps. So, they couldn't just scale them

to be more capable. This led to stage

two, which I think of as starting in the

summer of 2024, which is where they

shifted their attention to post

training. So now like we can't make the

underlying smarts of these LLMs

um better by making them bigger,

training them longer. So what we need to

do is try to get more useful stuff out

of these existing pre-trained LLMs. And

so the first approach they came up with

and we we saw this with the the alphabet

soup of models that were released

starting in the fall of 2024 01 03 nano

banana like all these type of names. The

first approach they tried was um telling

the models to think out loud. So instead

of uh just directly producing a

response, they post-train the models to

be like actually explain your thinking

and it was sort of a way because

remember it's auto reggressive. So as

the model sort of explains its thinking

that's always going back as input into

the model and it gives it more to work

off of in reaching an answer. So it

turned out if you had the model think

out loud you got slightly better on

certain types of benchmarks. So these

were the so-called reasoning models. Um,

but it was a bit of a wash because this

also made it more expensive to use the

models because it burned a lot more

tokens because the answers you it

produced a lot more tokens to get to the

answer you cared about. So it did better

but it was unclear like how much of that

we actually want to turn on for users.

Um, the second approach they used in the

second stage was post training. So now

if you have for example a lot of

examples of a particular type of

question prompts correct answers prompts

correct answers you could use those

combined with techniques out of

reinforcement learning to nudge the

existing pre-trained model to be better

on those type of tasks. So, we entered

this stage, stage two of of the sort of

post-training stage where because we

couldn't make these uh LLM brains

fundamentally smarter, we wanted to try

to tune them to get more performance out

of them uh on particular types of task.

This is when we began to see less of

just, hey, try this model and it's going

to blow blow your socks off and we

instead got lots of charts of inscrable

benchmarks. Look, the the chart is going

up on this alphabet soup benchmark

because, you know, you could post train

for particular benchmarks. It was less

obvious in a lot of use cases for the

regular user that like well the

underlying smart seems to be the same.

We then entered a stage three. I think

this started in the fall of 2025

where the LLM company said really the

big gains going forward is in the

applications that use the LLMs. Let's

make these applications smarter. So it's

not just how capable the LLM is. It's

like how capable is the programs that

are prompting the LLM. Let's make those

smarter. So we saw a lot of this effort

going into the programs that are called

coding agents that help computer

programmers edit and produce and plan

computer code. Now these type of agents

had been around for many years but they

got really serious a lot of the AI

companies especially uh last year coming

into the fall of last year and how do we

make the program so they weren't

changing really much the LLMs they did

some fine-tuning for uh programming but

really the big breakthroughs in coding

agents were in the programs that call

the LLMs and they figured out how can we

make these coding agents capable of

working with enterprise code bases so uh

not just for individuals vibe coding web

apps but something you could use if

you're a professional programmer in a

big company. All of that's tool

improvements.

Making sure that you're able to send

better prompts to the LLM. When you hear

about things like skill files and um

managing like hierarchies of agents,

this is all improvements in the programs

that use the LLM, not the none of this

is breakthroughs in the digital brain

itself. And so this is the stage that we

are in now is we're spending a lot more

time building smarter programs that sit

between us and the LLMs that they're

quering as their digital brain so that

it's in it in very particular domains it

is more useful.

So this all tells us right this is like

what Lun would tell you right I'm I'm

channeling Lun he would say once you

understand this reality you see that

this impression that LLM based AI has

been on this super fast like upper

trajectory of lots of fast advances is

pretty illusory

the fundamental improvements in the

underlying brain stopped a couple years

ago

what we saw was then a period of lots of

brag bragging about benchmarks doing

better but this was all about post-

training and now for the last four

months like all these improvements we've

been hearing is about the programs that

use the LLMs are being made smarter and

they're better fitting particular use

cases but there really hasn't been major

fundamental uh improvements in the

underlying smartness of the digital

brains which is why all the problems

like hallucinations and unreliability

persist. the brains are actually

incrementally improving either in narrow

areas um or in narrow ways. And it's

what we're building on top of them. This

creating an illusion of increasing

trajectory of artificial intelligence

when in reality we might just be in a

very longtail stage of now we're going

to do product market fit and actually

build do the work of building more

useful products on top of a mature

digital brain technology that's only

advancing at a very slow rate. That

would be Lun's argument. Uh therefore we

will find some good fits, but this is

not a technology that's on a trajectory

where it's going to be able to make

massive leaps in what it's actually able

to do.

All right. Um

so there you go. That would be the

argument for how we could have gotten

LLM progress so wrong. All right. Sub

question number three.

Let's follow through this thought

experiment. What would happen if Lun is

right about that? What what would we

expect then to happen in the near

future? Well, let's start with the the

window of the next one to three years.

If he is right, we would see a long tale

of applications based on existing LLMs

to begin to fill in. So, computer coding

agents have gotten more useful. We will

see other use cases like that that don't

exist now, but where people are really

experimenting to try to figure out

applications that are going to uh work

in other types of fields. So there'll be

sort of claude code moments in other

fields which I think will be useful and

exciting. Um the tool sets used in many

jobs will change but because we're now

just trying to like find areas where we

can build useful applications on top of

existing LLMs. these doomsday scenarios

like we we've been talking about on this

AI reality checks recently where

knowledge workers are going to have to

become uh pet massuses and then after

that they're going to have to cook the

pets on garbage can fires because

there's no money left the economy none

of those scenarios are are would unfold

based on LLMs in this current vision

there would be a big economic hit

because what we're going to if we've

shifted our attention to building better

applications on top of the LLM what

we're going to see is a lot more

companies get into that game and they're

going to say, "I don't want to pay for a

cutting edge frontier hyperscaled LLM.

It's too expensive. Let's look at

cheaper LLMs. Let's look at open source

LLMs. Let's look at LLMs that can fit on

chip."

We saw this already with the OpenClaw

framework, which allowed people to build

their own custom applications that use

LLMs to do personal assistant type

roles. And right away, people are like,

I don't want to pay all the money to use

Claude or GPT. And you saw an explosion

of interest in onchip machines and open

source machines. All this is going to be

I think good news for the consumers.

That means we could have more people

building these applications. There'll be

more variety of these applications and

they'll be cheaper. It's bad news for

the stock market because we've invested

depending on who you ask somewhere

between 400 to 600 billion dollars into

these LLM hyperscalers like OpenAI and

Anthropic.

That market's not going to support it.

So there's going to be a big crash. This

will probably temporarily slow down, if

this vision is correct, would

temporarily slow down AI progress

because investors are going to feel

burnt. All right. What's going to happen

now if we zoom out to like a a 3 to 10

year range? Um, that's roughly the range

in which the modular architecture

approach that Lacun is talking about

would reach maturity. That's what their

current CEO is saying. Again, it's it's

a research company now and they said

it'll be several years until we really

get the products that are ready for

market. If lacun is right, what we're

going to see is domain by domain, you're

going to have these uh very bespoke

train domain specific modular

architecture systems which if he's right

are going to be way more reliable and

more smart in the sense of like they do

the thing I asked them in a way that's

good and as good as like uh some of my

human employees and in a way that I can

actually trust. We're going to see a lot

more of that. what's promised with LLMs

we're going to see instead on that 3 to

3 to 10 year basis if lacun is right

because they're uh based on this mild

jar architecture I think these systems

will you know they'll be more reliable

um they're also going to be easier to

align

LLMs are so offuscated it's just like

here's 600 billion parameters in this

big box that we trained for a month on

all the text on the internet let's just

see what it does modular architectures

are way more alignable like you have

literally a critic module in there that

evaluates plans based on both a world

model and some sort of hard-coded value

system to say which of these do I like

better and you could just go in there

and sort of hardcode don't do these type

of plans you know uh really have a low

score for plans that lead to whatever

like a lot of variability in in outcome

or something like that you have more

direct knobs to turn so it does make

alignment more easier um they would also

be more economically efficient because

when you're when you have to train one

module long enough one model long enough

that could everything. It has to be huge

and it takes a huge amount of energy.

But when you're training different

modules in a domain specific system,

these can be much smaller. I like to

point out the example of a deep mind, a

Google DeepMind tool called Dreamer V3,

which can learn how to play video games

from scratch. It's it's famous for

figuring out how to find diamonds in

Minecraft. And it uses a modular

architecture um very similar to what

Lacun is proposing here. And we just

read a paper about it in my doctoral

seminar I'm teaching on super

intelligence right now. Dreamer v3 which

can play Minecraft it well better than

if you ask an LLM to do right it's

domain specific requires around 200

million parameters which is a factor of

10 or less than what you would get in a

standard LLM. It could be trained on a

single GPU chip and it could do this

domain way better than uh a frontier

language model which is significantly

longer and train significantly more

exhaustively. So there would be some

advantages here. There would also be

some there's a little bit of digital

lick around this world because uh way

more so than LLMs again these domain

specific models might actually have more

of a displacement capability. So we'd

have to keep an eye on them. All right,

conclusion. What do I think is going to

happen here? Well, you know, I don't

know, right? It's possible that there's

more performance breakthroughs to get on

LLMs and we're going to get more useful

tools. A gun to the head if I had to

predict,

you know, through my computer science

glasses, lacun's modular architecture,

it feels like that has to be the right

answer.

I I think of this doubling down on LLMs

is we're going to look back at this like

an economic mistake. It was the first

really promising new AI technology uh

widespread AI technology built on top of

deep learning and it did cool things.

But instead of stepping back and like

okay what will this be good for and what

types of domains might we want different

models we said no let's just raise half

a trillion dollars and just go all in on

everything textbased LLMs which are

trained on text and are made to produce

text all artificial intelligence will

run off of these things. I just think

when we zoom out on the 30-year scale,

we'll be like that was so naive. This

idea that like this was the only type of

model we need for artificial int. It's

super inefficient for like 99% of the

domains we want to use. It's great for

textbased domains and computer

programming kind of the planning is a

little suspect, but the code production

is okay. But we're going to make all

intelligence based off just massive LLMs

and there'll be like four of them, like

four companies that have like massive

ones and that's it. That this can't be

the right way to do it. So my my

computer science instincts say modular

architecture it just makes so much more

sense domain specificity differential

training of modules

you have much more alignment capability

they're much more economically feasible

like it just feels to me like that

probably is going to be the right answer

which means we're going to have to have

some bumpiness in the stock market

because I don't think that if this is

true the hyperscalers is now either they

have to pivot to those quick enough

before they run out of money or some of

them are going to go out of the business

and the others are going to have

collapse before they expand again. So I

think the modular architecture approach

will work better. I don't know if

Lacun's company's going to do it or not,

but I think that architecture it makes a

lot of sense to a lot of computer

scientists. Now, I hope they don't get

too much better because I'm much more I

can much more imagine a very trained

modular architecture AI digital brain

creating justified ick than I can

building these Python agent programs

that access some sort of massive LLM

somewhere. All right. So, yes, we'll

know. I think within a year we'll begin

to get a sense of which of these

trajectories is actually true. Um, I of

course will do my best to keep you

posted here on the AI reality check. All

right, that's enough computer science

talk for one day. Hopefully that made

sense. Hopefully that's useful.

Be back soon with another one of these

checks. And until then, remember, take

AI seriously, but not everything that's

written about it. Hey, if you like this

video, I think you'll really like this

one as well. Check it out.