Investors are reacting to growing concerns about AI’s potential for uncontrolled development, sparked by public resignations, failed AI experiments, and statements from top tech leaders like Sam Altman and Dario Amade urging a slowdown. This has led to a sharp drop in tech stock prices, particularly in chipmakers reliant on AI infrastructure. While the US government has not introduced federal AI regulations, industry leaders are pushing for voluntary pauses and tighter oversight, raising fears of global security risks such as AI-enabled bio-weapon development. Simultaneously, US supply chains are under strain due to tariffs, energy inflation, and rising demand for electronics parts from AI expansion. Foreign investment is shifting from US treasuries to equities, as the 10-year yield hits 5%, signaling weakening confidence in safe assets. In China, new travel restrictions targeting tech professionals reflect a broader strategy to protect intellectual property and maintain technological self-reliance amid US competition. These measures echo historical travel controls, though they operate within a more open system. Ultimately, the AI slowdown and geopolitical tensions are reshaping investment decisions, supply chains, and global tech competition, with long-term implications for innovation, financial markets, and national security.
Good morning from the Financial Times. Today is Tuesday, September 15th, and this is your
FT news briefing. Investors are not loving AI's existential crisis, and US supply chains
are struggling. Plus, China is restricting which of its citizens can leave the country.
I'm Sonya Hudson, and for Mark Filipino, and here's the news you need to start your day.
Tech stocks fell yesterday as markets reacted to ongoing talk of an AI apocalypse. Over the
weekend, AI pioneers Sam Altman and Elon Musk joined inthropic Stario Amade in calls to slow
down AI development. They're worried that humans could lose control of the technology. This
comes after an anthropic researcher quit publicly last week over concerns that the industry
is "gambling with our lives". FT columnist John Foley has been looking at the market implications
of a slowdown, and he argues there's a financial benefit too. He joins me now. Hi, John.
So let's start with why everyone is suddenly really worried about AI and also AI companies.
So the two things that have really caught people's attention. One is the online resignation
note from this researcher Jacob Cox and from an anthropic who said that he feels that neither
open AI nor anthropic, which obviously the two leading makers of AI models are acting responsibly.
Paying shortly after the incident in which an open AI experiment went a bit wrong and one
of its bots ended up hacking a company called Hugging Face, which has created all kinds
of soul searching about whether AI has reached the point where it's now doing things that
we didn't expect it to potentially causing a lot of damage. And Dario Amade, who's the
head of anthropic, he published a long essay this weekend in which he said it's time
to slow down the pace of AI development. So now everyone is asking the question of exactly
what that means, what it means to slow down, how they would do it, who would do it,
and whether it would really help. So Amade is worried that runaway AI could cost billions
of dollars and it could also endanger global security like if AI figured out how to build
bio weapons. Is the US government thinking about doing anything about this? That's where
a lot of these companies are based. So the US government has not shown much appetite
so far for passing federal regulation that limits the behavior of AI companies. And President
Trump has responded to the recent concerns about rogue AI by saying that he thinks that
this is a conspiracy that basically hands an advantage to China. Now what Amade is talking
about this weekend is essay is the sort of voluntary slowdown enhanced with some regulatory
measures and some kind of diplomatic measures. But it basically amounts to the company's
volunteering to be less effective at innovating, at least for the foreseeable future.
Now in response to all this text docs fell yesterday, chip makers were hit especially hard.
Should AI companies and their shareholders be worried? So the fall in share prices reflects
the idea that if these so-called frontier AI companies are not investing as heavily
in the new models, that there will be much less demand for data centers and for some
of the chip made by companies like Nvidia that are essential for the most sophisticated
AI models. So that's obviously bad for all the people who've been fighting to build data
centers on every available patch of earth all over the planet. All that said, for investors
and companies like Anthropic and Open AI, I have to believe that it is good if those
companies are less likely to cause some kind of AI Doomsday scenario. Obviously if we
have an AI Doomsday scenario and everyone's portfolio goes to zero, that's not good
for anyone. But also it means that those AI companies themselves are going to spend
a bit less on developing these models. They can focus on monetizing the models that they
have. And more of that would be great for an investor in Anthropic and also an Open AI.
So John, Anthropic and Open AI are both looking at IPOs. What could this turn of events
potentially do to those flotations? So I think Open AI has already seems to have delayed
its IPO. I think with any IPO, you want to reduce the number of variables in the company's
valuation. And if these companies are actively debating whether they need to slow down their
rate of progress. And we're actively debating with these companies, bring a risk of making
humans extinct. I think ideally you want to wait to price the shares until you know
a bit more about that. At the same time, they both need lots of money. They're both doing
very heavy investment. So I think the more they wait, the better for their investors,
but they really can't afford to wait forever.
John Fully is the head of the FT's Lacks column. Thanks, John. Thank you.
US manufacturers are stuck in a supply chain squeeze. It's pretty bad. What manufacturers
in the US are experiencing at the moment is just the price of everything going up considerably.
That's the FT's US economics correspondent, Miles McCormick.
The cost of what's known as intermediate goods, which is the stuff that manufacturers
buy in order to make other stuff is of about 12% since last year. And for the electronics
industry in particular, it's actually becoming increasingly hard to secure the parts that
they need to make things. That's because inflation is ravaging American industries. Miles
says there's three factors at play here. The first is Donald Trump's tariff regime,
which has really driven up the price of imported goods. The second is the war in Iran. Energy
prices have just become so much higher. And the third is the AI boom. The fact that companies
are scrambling to secure resources to build out data centers just means that in the electronic
sector, there's not as much parts available to go around. And Miles says American manufacturers
will only feel some relief once one of those three things cools off a bit.
In addition to those factors, if the Fed decides to hike interest rates as is widely expected
this week, that could help to tame inflation dandelion. But all of these are things that
if they do happen, it'll take a significant time for the cost effects to materialize.
So in the near term, at least, we're in for a bit of an inflationary picture.
Miles McCormick is the FT's US economics correspondent.
Foreign investors are now buying more US stocks than government bonds. It's a rare move
that shows the risk-free status of treasuries is under threat. The S&P 500 is on track
for a fourth-street year of double-digit gains because of AI investment. But things aren't
going so well for treasuries. The 10-year yield hit 5% yesterday. It's the world's most
important financial gauge, and this moves borrowing costs into unsettling territory. Yields
rise as bond prices fall. The 10-year yield briefly rose to 5% in 2023. But the last time
it consistently hovered above that level was in the lead-up to the global financial crisis.
Government bonds around the world have sold off this year because of inflation, growing
public debt, and blockbuster corporate bond issuance. The rise in treasuries yields
is likely to push up mortgage rates, which have jumped to 6.8% in recent weeks. It will
also put pressure on corporate America by making capital more expensive for businesses.
New exit laws are going into effect today in China on overseas travel for its citizens.
To step back toward Mao-era prohibitions. But today, the backdrop is tech competition
with the U.S. And the restrictions are part of an effort to secure advanced technology
and retain highly skilled workers. Jolay is the FT's Beijing Bureau chief, and he's
here to talk about it. Hi, Joe. Hi, Sonia.
So what exactly are these new restrictions?
Yeah, so these new restrictions are quite sweeping compared to what we've had in the past.
And one of the key ones is that now if you are accused of violating an export ban or
endangering national security in very broad language like that, you can be subject to an
exit ban in China in that they will stop you from leaving the country. And these
bands can range from anything from three months to indefinitely. So for example, you
might have a tech executive who they feel has extremely sensitive intellectual property
and they don't want that person going to Silicon Valley or somebody like that and maybe
doing deals and they can stop them. And we saw this with a deal when Manus and AI company
in China was sold to Meta. And after that deal, which the Chinese authorities didn't
like, they slept in exit ban on the management of Manus.
Now, I mentioned earlier this is sort of a step back towards Mao-era prohibitions. How
similar is this to some of the restrictions that we saw for decades in China before it
opened up?
Yeah.
it's really the principle here. I mean, obviously China's not at the stage where it was,
you know, in the Mauiro where no one could travel and few people could come in.
It's still much more open. In fact, there's a lot of visa free travel into China.
But the idea really is that the right to travel becomes, you know, a privilege that's
decided by the state for a large number of people. One group that we haven't mentioned yet is
government officials. All kinds of government officials, you know, have to surrender their
passports and when they want to travel, they have to ask for permission. That's already happening
in China. Joe, just taking a step back, what do you think these new rules tell us about how the
government is dealing with its competition with the West? Yeah, I think what we're seeing here is
as China moves up the ladder of technological advancement, it is becoming more and more sensitive
about passing that IP to the West and its answer to that so far has been to really strengthen
its laws on national security and we've seen a whole spate actually of what they call foreign
related law, which is all about trying to tighten up on investment, tighten up on supply chains.
We're seeing this sort of legal fortress being created in China as it as it competes with the West.
Joe Lay is the FT's Beijing bureau chief. Thanks, Joe. Thanks, Sonia.
You can read more on all these stories for free when you click the links in our show notes.
This has been your daily FT news briefing. Check back tomorrow for the latest business news.
Podcast Summary
Key Points:
Tech stocks, especially chip makers like Nvidia, fell due to concerns over an AI slowdown triggered by warnings from AI leaders Sam Altman and Dario Amade about runaway AI risks.
The resignation of an Anthropic researcher and a bot hacking Hugging Face sparked fears about AI's loss of control, prompting calls for voluntary slowing down of AI development.
Despite US government inaction on federal AI regulation, companies are advocating for self-imposed slowdowns and enhanced oversight to prevent potential global security threats.
US manufacturers face severe supply chain pressures from rising intermediate goods costs, driven by Trump's tariffs, the Iran war, energy inflation, and AI-driven demand for electronics parts.
Foreign investors are shifting from US government bonds to equities, as the 10-year Treasury yield rose to 5%, signaling growing risk aversion and inflationary pressures.
China has introduced strict new travel restrictions, allowing exit bans for citizens accused of endangering national security or violating export laws, especially in tech sectors.
These measures reflect a broader effort to protect intellectual property and retain skilled workers amid intensified tech competition with the United States.
A slowdown in AI innovation may reduce short-term investment demands but could improve financial stability for AI firms, potentially impacting their IPO timelines and investor valuations.
Summary:
Investors are reacting to growing concerns about AI’s potential for uncontrolled development, sparked by public resignations, failed AI experiments, and statements from top tech leaders like Sam Altman and Dario Amade urging a slowdown. This has led to a sharp drop in tech stock prices, particularly in chipmakers reliant on AI infrastructure. While the US government has not introduced federal AI regulations, industry leaders are pushing for voluntary pauses and tighter oversight, raising fears of global security risks such as AI-enabled bio-weapon development.
Simultaneously, US supply chains are under strain due to tariffs, energy inflation, and rising demand for electronics parts from AI expansion. Foreign investment is shifting from US treasuries to equities, as the 10-year yield hits 5%, signaling weakening confidence in safe assets. In China, new travel restrictions targeting tech professionals reflect a broader strategy to protect intellectual property and maintain technological self-reliance amid US competition.
These measures echo historical travel controls, though they operate within a more open system. Ultimately, the AI slowdown and geopolitical tensions are reshaping investment decisions, supply chains, and global tech competition, with long-term implications for innovation, financial markets, and national security.
FAQs
Tech stocks are declining due to growing concerns about AI's potential to become uncontrollable, prompting leaders like Sam Altman and Dario Amade to call for a slowdown in AI development to prevent harmful outcomes.
AI leaders are worried that rapid development could lead to 'runaway AI' capable of causing billions in damage or threatening global security, such as by building bio-weapons or losing human control.
The US government has not introduced federal regulations to limit AI companies' behavior. President Trump has dismissed AI risks as a conspiracy benefiting China, while companies are pushing for voluntary slowdowns.
A slowdown in AI development may reduce investment in new models, leading to less demand for chips and data centers. This could hurt short-term growth but improve long-term stability by reducing existential risks.
AI companies like OpenAI may delay their IPOs until they can demonstrate greater stability and responsibility, as uncertainty around AI safety could make valuations risky for investors.
US manufacturers face rising costs due to Trump’s tariffs, higher energy prices from the Iran war, and AI-driven demand for electronics parts, all of which strain the availability of intermediate goods.
Chat with AI
Loading...
Pro features
Go deeper with this episode
Unlock creator-grade tools that turn any transcript into show notes and subtitle files.