... sentence of your column, "Slow down should be a full stop."

stop."Yeah.

I love that you pointed to that word pacing, 'cause I think it's very interesting that he used that and slowed down, 'cause pacing is different to slow down.

Pacing doesn't mean slow.

It means kind of working with a rhythm or a strategy or a kind of plan to get to or to manage a certain path to get to a particular goal.

And I thought it was very interesting that the CFO of Anthropic posted an, um, Dario Amodei's essay on LinkedIn, and he didn't use the word slow down at all.

He just used the word pacing in his own explanation of it.

Uh, the likelihood of that happening?

Why are so many other policymakers pushing ahead with measures like this, even if the evidence we have so far isn't terribly strong about their effectiveness?

I think it's because this is what parents want, and it really shouldn't surprise...

And I think this is sort of a long time coming if, if we think back to the Cambridge Analytica scandal with Facebook.

This was probably the real turning point for the general public's perception of technology companies.

And ever since then, as these companies have grown bigger and the users on their platforms have gone into the billions of people, and there's been more reporting in the press and whistleblowers showing the harmful side effects of being on social media, the toxic content, the going down rabbit holes, the misinformation and conspiracy theories and, you know, teenage girls who feel worse about themselves when they're on Instagram.

That was a very famous study that came out of Facebook's own research, and we know about that because of the whistleblower, Frances Haugen, um, who, um, released a whole trove of documents.

So all this over the last, you know, five to 10 years has been slowly building this gradual kind of understanding that actually while these platforms are fun to use, and they are addictive, and people, you know, scratch this kind of mental itch when they get to go on their Instagram or, or Snap or whatever it is, there is this toxic side effect.

So besides the age-related bans, what other methods are regulators looking at to try to reduce the harm from social media?

How important is AI going to be for the future of the world?

I think it's going to be very important, but I think it's going to be hard at first to see the impact.

And I think one way to think about it is the Internet.

When people in the 50s and 60s, 1950s and 1960s tried to imagine what the future would look like.

They imagined robot butlers and flying cars.

And of course, we don't have that.

But we have something that was just as monumental, if not more so, which was the web.

How is Europe doing? Is Europe lagging behind everything or are there some encouraging developments there?

What kinds of products, I suppose, first of all, are we actually starting to see being used? And then maybe we'll think about what that means for employment in a minute.

Well, most people are using chatbots already, right? I think it's something like 800 million to 900 million people use ChatGPT on a weekly basis.

That's something like 10% of the global population every week are using these tools.

But the most recent updates have been agentic AI.

Of course, you might have heard that term agent a lot last year.

It was very much hyped.

Unfortunately, that didn't have much to show for it.

How far away is the future where artificial intelligence does actually make workers redundant? Because we're starting to see bits of reporting around that and bits of data in Britain.

What kinds of products, I suppose, first of all, are we actually starting to see being used? And then maybe we'll think about what that means for employment in a minute.

Well, most people are using chatbots already, right? I think it's something like 800 million to 900 million people use ChatGPT on a weekly basis.

That's something like 10% of the global population every week are using these tools.

Wow.

But the most recent updates have been agentic AI.

Of course, you might have heard that term agent a lot last year.

It was very much hyped.

How far away is the future where artificial intelligence does actually make workers redundant? Because we're starting to see bits of reporting around that and bits of data in Britain.

What kinds of products, I suppose, first of all, are we actually starting to see being used? And then maybe we'll think about what that means for employment in a minute.

Well, most people are using chatbots already, right? I think it's something like 800 million to 900 million people use ChatGPT on a weekly basis.

That's something like 10% of the global population every week are using these tools.

But the most recent updates have been agentic AI.

Of course, you might have heard that term agent a lot last year.

It was very much hyped.

Unfortunately, that didn't have much to show for it.

How far away is the future where artificial intelligence does actually make workers redundant? Because we're starting to see bits of reporting around that and bits of data in Britain.

These concerns you share, Parmi, good morning.

Hi, good morning.

Yeah, I mean, we've obviously been seeing big tech companies spending a lot on AI and the infrastructure to build AI over the last 12 to 24 months.

So this is nothing new.

But I think what we're really starting to see and what really intrigues me about this is this kind of almost an identity shift Because if you think about why they're so valuable in the first place, it's that their asset light, right? Like they have a lot of staff and they maybe have some data centers, but then they build this incredibly valuable software and they have incredible margins.

Now they're becoming these kind of capital intensive infrastructure.

They're almost like telecoms now with all this capex spending.

Are we seeing the same kind of reaction in the market to this latest tool from Anthropic? Yeah.

Do you see that level of capex, as Tom talked about, continuing next year? And as we see this roll-out of AI, what are the potential bear traps there waiting down the road for the administration or the economy?

Well, I think you're absolutely right.

In a lot of ways, the AI boom has given the Trump administration a lot of cover because it's just made it look like everything is going so well.

And in aggregate, the Magnificent Seven tech stocks have seen their market valuations increase by trillions of dollars just in the last three years since ChatGBT came out.

I personally think there is going to be some kind of correction in 2026.

And for all the reasons that people are saying, which is that there is very high risk of near term monetization not happening.

In other words, there's been this huge build out of infrastructure anticipating all this demand that probably will come in eventually, but it's just not gonna come in fast enough to meet the expectations of investors.

What should we be watching on that? What are the key debates we should be looking out for next year if we're kind of concerned about the form that AI is gonna take?

What impact is that going to have on the job market, on education and such like?

Well, first of all, I'm really interested in that point you made and the kind of examples you gave about not having to remember certain sums or directions.

I suppose now people are using GPS.

And that doesn't seem like a particularly big price to pay, but losing the ability or the desire to write an email, to write a response, to write.

I feel like that's a more fundamental issue.

Because language is kind of core to what makes us humans and how we express ourselves and our characters and our personalities.

So there's that point on the job market.

But they're

As did Sam Altman.

As did Sam Altman.

So on the other side of the world, in St. Louis, Sam grew up in this middle class household, oldest of four kids.

And he actually came out as gay to his entire student body and was very proud and open about that.

And also very mission oriented and very good at connecting with figures of authority in his high school to help get across whatever point he wanted to get across, like support for the LGBT community.

in his school.

When he left to study computer science at Stanford, he then got quite lucky and found out about the so-called, it's called Y Combinator.

It's fair to say, isn't it?

Parmi, since you have just written this book, and we have heard a lot about deep seek in the last six months or so, if the president wants the US to win the race in AI, is the US still the leader in this technology? Or did deep seek show us that China had already caught up?

Well, it depends how you define leader.

But I would say when it comes to AI capabilities, the models and how good they are at reasoning and math and coding, then yes, U.S. models are still ahead of the ones coming out of China.

The thing is, they're not ahead by much.

It's a very narrow lead.

Now, when it comes to actual adoption of AI services, I would say that in China, it's much more sort of readily available.

People use AI much more in their daily lives.

Oh, that's always

And when was the moment you thought, there's a book here and I'm going to write it? Because you do have quite a busy day job as well.

Yeah.

So I had been writing about AI probably for about seven years.

First with Forbes, actually longer than seven years.

First with Forbes, then with the Wall Street Journal.

And I continued writing about it.

with Bloomberg and there have been a few times when editors have told me you know this is kind of like maybe you could broaden your beat a little bit but I really wanted to stick with it because I am fascinated by the people trying to build these things and the reason I wanted to write the book was because when ChatGPT came out I knew that behind the scenes was this goal to not just build a chatbot, but to build a god, to build AGI, which is Artificial General Intelligence, or this AI that surpasses human intelligence.

But yeah, well, let's talk a bit about the kind of obviously not exit to Google, but the sort of how they how they fold in with Google, because it basically comes down partly to access to computing power, doesn't it?

did Sam Altman.

As did Sam Altman.

So on the other side of the world, in St. Louis, Sam grew up in this middle class household, oldest of four kids.

And he actually came out as gay to his entire student body and was very proud and open about that.

And also very mission oriented and very good at connecting with figures of authority in his high school to help get across racism.

whatever point he wanted to get across, like support for the LGBT community in his school.

When he left to study computer science at Stanford, he then got quite lucky and found out about the so-called Y Combinator program.

It's fair to say, isn't it?

So why is this taking place now then?

Well, as I understand it, this is an annual event.

And so there was one in 2023.

Then there was another one in Seoul, South Korea last year.

And this year, we've got one in Paris.

The reason these have started is, again, just because of these rapid advancements.

You ChatGPT came out two years ago, the rate at which AI models have been improving is even taking scientists by surprise.

Could things get a bit tense given that everybody is jostling to be at the forefront of this AI race?

So why is this taking place now then?

Well, as I understand it, this is an annual event, and so there was one in 2023, then there was another one in Seoul, South Korea last year, and this year we've got one in Paris.

The reason these have started is, again, just because of these rapid advancements.

You know, since ChatGPT came out, two years ago, the rate at which AI models have been improving is even taking scientists by surprise.

Some of the academics studying this stuff will read one paper.

By the time they finished reading it, it's sort of out of date because Anthropic or OpenAI or Google have come up with something more incredible, like a video generation tool.

And for businesses, investors and policymakers, it's been very tricky just to keep up, whether it's to try and plug these systems into your businesses or try and put the standards in place to keep them safe.

Could things get a bit tense given that everybody is jostling to be at the forefront of this AI race?