Why are we trying to build AI systems
that are duplicative of humans? They
said that we should be building AGI, so
we say that we should be building AGI,
but I would like to ask like why are we
doing that? Why is it that we are
building a technology that is ultimately
designed to replace and automate people
away? We should be building technology
and the purpose of technology throughout
history has been to improve human
flourishing, not to replace people. Is
this the right goal? What like why are
we doing this? Why can't we just build
AI systems, which are systems that have
nothing to do with the statistical
engines that they're trying to build to
duplicate the human brain?
>> That might be the most uncomfortable
question I've heard about AI this year.
Not can AI become smarter than humans,
not will AI destroy humanity, but
something much simpler. Why are we
trying to build machines that replace
people instead of helping people? Think
about it for a second. When cars were
invented, they helped us travel faster.
When smartphones became common, they
connected us to the whole world.
Technology has always been about giving
humans more power, but for the first
time, millions of people are asking a
completely different question. What
happens when technology starts replacing
the person instead of just helping the
person? And that's exactly why I wanted
to react to this podcast. A 2-hour
conversation between Steven Bartlett and
journalist Karen Hao, a woman who spent
8 years and over 300 interviews
investigating what's really happening
inside companies like OpenAI.
And I need to be honest with you. Some
of what she says made me angry. Some of
it made me question things I use every
day, and one part about a small town in
Tennessee genuinely disturbed me.
Let's walk through it and give my honest
opinion about it.
I want to give you some background first
because you can't understand the fight
without understanding the field.
>> So, this was back in 1956
and [music] there were a group of
scientists that gathered at Dartmouth
University to start a new discipline, a
scientific discipline, to try and chase
an ambition. And specifically, an
assistant professor at Dartmouth
University, John McCarthy, decided to
name this discipline artificial
intelligence.
This was not the first name that he
tried. The previous year, he tried to
name it Automata Studies. And the reason
why some of his colleagues were
concerned about this name was because it
pegged the idea of this discipline to
recreating human intelligence.
>> His colleagues were deeply concerned
about the term artificial intelligence
because it tied the field to the goal of
recreating human intelligence. Both then
and now, there is absolutely no
scientific consensus on what human
intelligence actually is. There is no
single definition from biology,
psychology, or neurology, which means,
as Karen points out, when today's AI
companies talk about building AGI,
artificial general intelligence, they're
chasing a finish line that literally
doesn't have fixed coordinates.
>> Well,
there are no goalposts for this field,
and there are no goalposts for the
industry when they say that they are
ultimately trying to recreate AI systems
that would be as smart as humans. How do
we even define what that means? And when
are we going to get there if we don't
know how to define the destination?
>> Karen shows how OpenAI has redefined AGI
depending on who's in the room. When Sam
Altman talks to Congress, AGI is the
thing that will cure cancer, solve
climate change, cure poverty. When he's
selling to regular customers like us,
it's the most amazing digital assistant
you'll ever have. When Microsoft was
writing the investment contract, AGI was
legally defined as a system that
generates $100 billion in revenue.
>> They can use it however they want to,
and they can define and redefine it
based on what is convenient for them.
So, in OpenAI's history, it has defined
and redefined it many times. When Sam
Altman is talking with Congress, AGI is
a system that's going to cure cancer,
solve climate change, cure poverty. When
he's talking with consumers that he's
trying to sell his products to, it's the
most amazing digital assistant that
you're going to ever going to have. When
he was talking with Microsoft, you know,
in the deal that OpenAI and Microsoft
struck where Microsoft invested in the
company, it was defined as a system that
will generate a hundred billion dollars
of revenue. And on OpenAI's own website,
they define it as highly autonomous
systems that outperform humans in most
economically valuable work. This is like
not a coherent vision of one technology.
These are very different definitions
that are spoken out loud to the audience
that needs to be mobilized to ward off
regulation or get more consumer buy-in
into the the industry's quest.
>> Read that again. Same three letters,
completely different meaning depending
on who needs convincing. I don't think
that's a coincidence. I think that's a
strategy.
Here's the clip that genuinely made me
sit up.
>> In 2015,
in a blog post that Sam Altman wrote
before OpenAI was officially announced,
he explicitly outlined the existential
risk by saying, "Development of
superhuman machine intelligence is
probably the greatest threat to the
continued existence of humanity. There
are other threats that I think are more
certain to happen, for example, an
engineered virus,
but AI is probably the most likely way
to destroy everything."
>> Back in 2015, before OpenAI officially
existed, Sam Altman wrote a blog post
that said, and I'm paraphrasing
carefully here, that a superhuman AI was
probably the single biggest threat to
humanity's survival, bigger than almost
anything else.
Karen's reporting shows that Altman
wrote that specific line right around
the time he was trying to convince Elon
Musk to co-found OpenAI with him. And
Musk, at that exact same time, was doing
famous talks, including one at MIT,
where he compared building advanced AI
to summoning the demon.
>> And so, in that blog post, if you look
at the the language that Altman uses
side by side with the language that Musk
was using at the time, it mirrors the
all the things that Musk was saying.
>> Yeah.
>> years ago Musk was going on podcasts
saying, tweeting, whatever, that the
greatest existential risk to humanity
was AI.
>> Musk did end up co-founding OpenAI. He
also ended up leaving, and by his own
account, feeling used. The two are now
suing each other.
But this next part is where Karen's
reporting feels like something you need
to actually know before you open your
chatbot app again.
>> And that is why I call these companies
empires of AI.
>> What do you mean by an imperial agenda?
What What does that term mean?
>> Empire is the only metaphor that I've
[music] ever found to fully encapsulate
all of the dimensions of what these
companies do,
and the scale that they operate, and
what motivates them to do what they do.
And there are many parallels that you
see between what I call the empires of
AI and the empires of old.
>> Karen's whole book rests on one bold
claim: These are not just companies. She
calls them empires of AI, and she lays
out four specific parallels to empires
of old. One, resource grabbing.
She says these companies lay claim to
resources that aren't theirs, the
personal data of ordinary people, the
intellectual property of artists and
writers, and land for building massive
computing facilities. Two, labor
exploitation.
Hundreds of thousands of workers around
the world, including in the US, are
contracted to help build these systems,
often in ways that quietly strip away
labor protections once the tools are
deployed.
Three, controlling knowledge.
These corporate empires project the
false narrative that they are the only
ones pure enough or smart enough to
understand and guide AI safely. They
actively gaslight the public and policy
makers. Even worse, they spend hundreds
of millions of dollars to kill public
safety legislation and ruthlessly censor
any internal researchers whose findings
threaten their corporate profits.
Take the case of Dr. Timnit Gebru at
Google.
She was hired to lead an ethical AI
team. But when she co-authored a
brilliant paper highlighting the severe
biases and environmental harms of large
language models, Google executives tried
to suppress it and ultimately fired her
and her co-lead, Margaret Mitchell.
Four, myth-making. This is the one that
got me. Karen says these companies need
the public to believe two things at
once.
That AI could be catastrophic and that
it could bring utopia. Why both? Because
if it's dangerous, only they should be
trusted to control it. And if it's
utopian, we should hand them our trust,
our data, and our money to get there
faster. Here's a real quote she brings
up from 2017.
>> the record on podcasts in their blog
posts to see how it's evolved over time.
And Dario, who was the former VP of
research at OpenAI, and has now moved on
to Anthropic, who are taking a slightly
different approach to developing AI,
said back in 2017 while he was still at
OpenAI, that I think at the extreme end
is the Nick Bostrom style of fear that
an AGI could destroy humanity. I can't
see any reason in principle why that
couldn't happen. My chance that
something goes really quite
catastrophically wrong on the scale of
human civilization
might be somewhere between 10% and 25%.
>> 10 to 25%. Read that number again.
That's not a fringe blogger. That's a
man now running one of the most funded
AI labs on Earth.
I thought the empire argument was the
most shocking part of this conversation.
I was wrong. The next section about a
small community in Tennessee is the part
that actually stayed with me overnight.
We talk about AI like it lives in the
cloud, like it's weightless. It is not.
>> OpenAI's largest, one of its largest
data center projects, is being built in
Abilene, Texas as part of the Stargate
initiative, which was an effort
announced at the beginning of Trump's
second administration to spend $500
billion on AI computing infrastructure.
That one would be the size of Central
Park.
And it would run a million computer
chips, and it would
require the power of more than 20% of
New York City.
>> Then there's a meta facility being built
in Louisiana that's reportedly four
times the size of that Texas one, using
about half the average power demand of
the entire city of New York.
Karen talks about Memphis, Tennessee,
where Elon Musk built Colossus, the
supercomputer used to train Grok,
powered partly by 35 methane gas
turbines.
The community living there, a working
class, reportedly wasn't formally told
they'd be hosting this facility.
According to Karen, residents first
realized something had changed when they
smelled what seemed like a gas leak
inside their own homes.
That community, she says, already has
some of the highest rates of asthma and
respiratory illness in the area.
Now add industrial-scale gas turbines
running next door, 24 hours a day, to
power a machine most of those residents
will never directly benefit from.
And in a cruel twist, the same wave of
automation this facility helps build is
also the thing most likely to replace
the jobs many of them do.
I want to be fair here, because I think
it matters. AI data centers also create
construction jobs, local tax revenue,
and in some cases genuinely useful
technology for medicine and science.
This isn't a simple villain story.
Karen says these companies pick some of
the most vulnerable communities for
these projects. And when you look at
where facilities like this actually
land, it's hard to call that a
coincidence, either.
Now, here's the part where you might
expect me to just tell you to log off
and boycott AI forever.
That's not what Karen says. Her answer
is more interesting and honestly more
useful.
Karen's suggestion isn't delete every AI
app. It's smarter than that. She calls
all of us data donors, meaning every
time we use these tools, we're feeding
the very systems these companies need to
keep growing. Her point, you have more
leverage than you think. She lists a few
concrete things, and I want to walk
through them simply. Withhold what you
can. Artists and writers suing these
companies over their work are, in her
words, trying to withhold the resource
these companies need most, original
human data. Get loud locally.
>> And that's what we've seen with all of
these people that are now literally
protesting the streets against data
centers and having an enormous effect,
by the way, actually stalling data
center projects and also completely
banning data centers from being
developed in their localities.
>> She points out that data centers are
popping up in ordinary towns right now,
and in several places, local communities
have already succeeded in stalling or
outright banning new data center
projects through public pressure. Push
at your school or workplace.
If your company or university is
drafting an AI adoption policy, that
conversation is one small piece of a
much bigger puzzle, and your voice in
that room genuinely counts. Support
alternatives.
Karen says there's research showing the
same useful AI capabilities can be built
with far more efficient methods and far
less resource consumption, what she
calls the bicycles of AI as opposed to
the resource-hungry rockets.
She also cites a striking number.
>> 80% of Americans in the most recent poll
think that the AI industry need to be
regulated.
>> Yeah, I see.
>> When was the last time that 80% of
Americans were on the same side of an
issue?
>> A shift she says happened largely
because of stories like the ones in her
book reaching the public. I want to end
this fairly because Karen herself does.
Late in the conversation, Steven pushes
back a little saying that as an
entrepreneur and investor, he genuinely
finds AI incredible. Safer driving,
tools that help him create things he
couldn't before. Real value for over a
billion users. And Karen doesn't dismiss
that. Her answer is basically, you can
hold both truths at once.
AI can be genuinely useful and the
current way it's being built and rushed
to market can be genuinely harmful.
Those aren't contradictions. They're two
honest observations about the same
industry. That's actually the most
balanced moment in the whole two hours.
And it's the note I want to end on, too.
So, here's my question for you and I
genuinely want to know. Knowing what you
now know about where the water for your
chatbot's training run might be coming
from, does it change how you'll use
these tools tomorrow? Not never. I'm not
asking you to swear off technology. Just
pay attention. Ask questions.
Support the version of AI that doesn't
need a small town in Tennessee to smell
gas in their living room to make it
work. If this made you think
differently, even a little, share it
with one person who uses AI every day
without ever asking these questions.
That's genuinely how change like this
starts.