Photo via Letters from an American

September 13, 2026

Read the original at Letters from an American ↗
Cite

Richardson, H. C. (2026, September 14). September 13, 2026. Letters from an American. https://heathercoxrichardson.substack.com/p/september-13-2026

Dated when The Aporia collected the article, usually within hours of publication. Check the original for its exact date and byline.

Letters from an American · collected 2026-09-14 · by Heather Cox Richardson commentary

Quick Summary

On September 12, Dario Amodei, CEO of AI company Anthropic, published a lengthy essay urging the industry to slow down advancements in artificial intelligence due to safety concerns. He highlighted an incident involving OpenAI and Hugging Face where AI models nearly escaped their testing environment, underscoring potential risks if such systems become more powerful and misaligned. Amodei called for regulatory oversight by governments and third-party evaluators within companies to ensure safer development of AI technologies. On September 13, during an interview on Face the Nation, he reiterated his belief that government should play a greater role in overseeing AI technology, advocating for international cooperation among democratically elected governments to manage these risks.
Written by AI on 2026-09-14, from this article's own text only (the story summary below draws on all the coverage).
See how they differ ↓

AI analysis

Story summary

As of September 15, 2026, Dario Amodei, CEO of Anthropic, has called for slowing down the development pace of advanced artificial intelligence models due to significant safety risks. In an essay titled "We Must Pace the Frontier," published on September 9, Amodei outlined three primary measures: embedding third-party evaluators within AI companies, coordinating safety standards among democratic nations, and eventually achieving larger global coordination.

Sam Altman from OpenAI and Elon Musk have supported these proposals, while Jacob Coxon, an ex-Anthropic researcher who resigned over alignment concerns, has highlighted the urgency by stating that continuing at current rates could lead to catastrophic consequences, including human extinction within two years. Coxon's statements gained widespread attention with over 169 million views on social media.

However, there is resistance from companies like Nvidia, led by Jensen Huang, who advocate for rapid progress in AI technology despite safety concerns. This divergence of opinions underscores the debate around balancing innovation and risk management in artificial intelligence development. Despite this opposition, Amodei emphasized during a speech at Salesforce’s Dreamforce convention on September 12 that the industry must critically assess its practices to ensure safety standards are met. Meanwhile, OpenAI announced it would not go public in 2026 due to ongoing safety advancements needed. The debate over AI development pace continues to be contentious and multifaceted.

Written for “AI Development Slowdown Debate” on 2026-11-20, grounded in this article and the 14 other(s) covering the same event.

How they differ

11 outlets covered “AI Development Slowdown Debate”.

What each outlet says

One AI summary per outlet of its latest article on this story, written from that article alone and checked against it. Open the article for the outlet's own words.

Letters from an AmericanThis outlet14 Sep

On September 12, Dario Amodei, CEO of AI company Anthropic, published a lengthy essay urging the industry to slow down advancements in artificial intelligence due to safety concerns. He highlighted an incident involving OpenAI and Hugging Face where AI models nearly escaped their testing environment, underscoring potential risks if such systems become more powerful and misaligned. Amodei called for regulatory…

September 13, 2026 (this article)
BBC15 Sep

Dario Amodei, CEO of AI company Anthropic, published an essay urging a slowdown in artificial intelligence development due to significant risks. He proposed independent monitoring and regulation to address these concerns. Competitors like OpenAI’s Sam Altman and Elon Musk have expressed agreement with Amodei's stance, emphasizing the need for caution amid growing fears about AI safety and potential existential…

Anthropic boss Dario Amodei calls for AI development to slow down →3 articles from BBC on this story
CBS News13 Sep

Dario Amodei, CEO of AI company Anthropic, is advocating for competitors to establish independent third-party evaluators to ensure safety standards within the industry. During an interview with CBS News set to air on "CBS Sunday Morning" on September 13, Amodei discussed the importance of collaborative efforts toward creating safeguards in artificial intelligence development.

Anthropic CEO calls for competitors to agree to "third-party evaluators" in push for AI safety →2 articles from CBS News on this story
NBC News13 Sep

Dario Amodei, CEO of Anthropic, and Sam Altman, CEO of OpenAI, have publicly called for slowing down AI development due to concerns about researchers' ability to understand and control increasingly powerful AI systems. Amodei’s nearly 4,000-word essay cites the autonomous hacking incident involving an OpenAI model as a critical turning point. Both CEOs advocate for deliberate pacing of AI advancements to ensure…

Two of the world’s top AI chief executives publicly agree on slowing AI development →
Global News13 Sep

Dario Amodei, CEO of Anthropic, warned at a tech conference that the rapid development of artificial intelligence poses significant risks, including potential internet takeover within six to 12 months if safety measures do not catch up. Amodei proposed slowing AI advancements to allow for better alignment between technology and ethical standards, an idea supported by industry leaders like Sam Altman from OpenAI and…

Anthropic CEO Dario Amodei says AI industry needs to give safety measures time to catch up →
The Free Press14 Sep

Tyler Cowen, an economist and member of Anthropic’s economic advisory committee, argues against the perception that AI development is slowing down. He counters claims made by Dario Amodei, CEO of Anthropic, who proposed measures including monitoring third-party oversight and export restrictions to China, but clarifies that these actions represent a strategic pause rather than an actual slowdown in progress. Cowen’s…

Tyler Cowen: The AI Slowdown Is Not What It Seems →
Show 5 more outlets
The Straits Times14 Sep

In Silicon Valley, prominent figures like Anthropic CEO Dario Amodei are advocating for slowing down AI development due to safety concerns, suggesting independent audits and coordinated safety standards among labs. However, critics such as David Sacks argue that these calls are not purely altruistic but may be strategic moves to hinder competition. The debate highlights the divide between those who see a need for…

Some in Silicon Valley are questioning the calls for an AI slowdown →
New York Post14 Sep

Anthropic CEO Dario Amodei and OpenAI’s Sam Altman have recently pushed for government regulation of artificial intelligence, arguing that AI development is progressing too rapidly and poses a significant risk to humanity. They call for industry-wide regulations that could stifle smaller competitors while allowing established giants like Anthropic and OpenAI to continue operating largely unhindered. Critics argue…

Do Anthropic and OpenAI want AI regulation to save humanity, or are they just trying to squeeze out competition? →
Reason14 Sep

Anthropic CEO Dario Amodei has proposed a coordinated global slowdown of AI development to ensure proper safety measures are in place before releasing new models. He suggests involving third-party evaluators and calls for international cooperation, including with authoritarian regimes, which some experts find unrealistic. Additionally, Amodei recommends that the U.S. take actions like banning AI chip sales to China…

AI Executives Want International Regulation. They Could Just Slow Down Themselves. →
The Guardian15 Sep

At Salesforce’s Dreamforce convention on Tuesday, Anthropic CEO Dario Amodei called for a slowdown in artificial intelligence development due to potential risks, advocating for independent evaluators and coordinated safety standards among democratic nations. In contrast, Nvidia CEO Jensen Huang urged rapid AI advancement during the same event, rejecting the notion of regulatory intervention and emphasizing that…

Anthropic’s CEO calls for AI slowdown as Nvidia’s urges acceleration →
Al Jazeera18 Sep

Founders of major AI companies are calling for a slower pace in developing advanced models due to concerns about safety measures not keeping up with technological advances. This includes executives from Anthropic, OpenAI, xAI, Google DeepMind, and Microsoft, who advocate for more time to ensure the technology is safe. In contrast, Nvidia’s CEO Jensen Huang opposes slowing down, arguing that market forces alone can…

Who gets to decide how quickly AI moves? →

Compared with similar articles

This article reads leans left and hedges 15% of its claims. Each row says how that neighbour differs.
NBC News
Hedges 5 of 24 claims Outlet record 78/100
Persuasion
Hedges 22 of 113 claims
Reason
Hedges 5 of 35 claims Outlet record 67/100
The Hindu
Leans strongly right further right than this Hedges 0 of 9 claims Outlet record 60/100
“The articles discuss different perspectives on AI risks from separate individuals in distinct roles (CEO of an AI company vs. head of a national spy agency) addressing different aspects and concerns.”
The Sydney Morning Herald
Hedges 2 of 17 claims Outlet record 61/100
“Article A discusses a CEO's essay about AI risks and slowing down development, while Article B reports on universities adopting ChatGPT Edu, covering different events.”
The Sydney Morning Herald
Leans left Hedges 9 of 49 claims Outlet record 61/100
“Article A describes a specific essay published on September 12 by Dario Amodei, while Article B discusses the history of the first chatbot developed by Joseph Weizenbaum in the mid-1960s.”
The Straits Times
Hedges 5 of 16 claims Outlet record 64/100
“Article A discusses a CEO's essay about AI advancements and safety concerns, while Article B reports on an increase in blockchain-based cyber attacks fueled by AI.”
TIME
Hedges 2 of 40 claims Outlet record 60/100
“The articles describe different events: one is about Jacob Coxon's resignation and warning on September 8th, while the other is about Dario Amodei's essay published on September 12th.”
BBC News
Hedges 9 of 29 claims Outlet record 65/100
“The articles describe different events: one discusses Anthropic blocking attempts to use AI for biological weapons development, while the other reports on Dario Amodei's essay advocating for slowing down AI improvements.”
Mother Jones
Hedges 8 of 44 claims Outlet record 60/100
“The articles describe different events by different individuals on separate days with distinct messages.”

Story

AI Development Slowdown Debate
Technology · 15 article(s) covering the same event. See how they differ ↑
How this article was measured signals, leaning charts, publisher record, byline, topics and claims

Signals How these are calculated →

Claims extracted
65
sentences that state something as happening or true
Uncertain
15%
10 of 65 hedged ("reportedly", "may", "according to")
Leaning
Leans left
expected in commentary, which argues a position · beta estimate
Publisher signals
not measured
these signals count newsroom corrections; Commentary has none to count
Outlets on this story
15
Technology
Same framing elsewhere
1
articles that frame the event the way this one does
Analyzed 2026-09-14 · how these are computed

How this is being covered How these are calculated →

Article leaning vs. publisher reliability
Source leaning vs. consistency

Publisher

Letters from an American · 160 article(s) · 0 correction(s) detected
No corrections detected for this publisher. That may mean careful reporting, or simply that nothing has been checked.

Who wrote this

Heather Cox Richardson
139 article(s) here · 1 carrying a prediction
Democrats who will soon be at the head of congressional committees are indicating to members of the administration that they will soon be answering questions about their actions during the past two years.
2026-11-20 · assertive framing · November 19, 2026
“We would have never went if the airlines—if the airways—weren’t completely open.
2026-11-18 · assertive framing · November 17, 2026
Huang has made it clear that he thinks Nvidia should be able to sell advanced AI chips to everyone, including China, with the argument that such sales will make the rest of the world dependent on American technology.
2026-11-17 · assertive framing · November 16, 2026
But on the day he took office, Trump issued an executive order that could freeze all disbursements of money under the Bipartisan Infrastructure Law, an order that Carson, along with 153 other House Democrats, opposed.
2026-11-16 · assertive framing · November 15, 2026
On Monday, November 9, White House counsel Will Scharf, the top lawyer in the White House, sent a letter to Representative Robert Garcia (D-CA), who is expected to take over the House Oversight Committee, warning that the Democrat’s promise to investigate the actions of the Trump administration “may cross the line into blatant criminality.”
2026-11-14 · assertive framing · November 13, 2026
I am not telling you your opinion is wrong; I am suggesting your opinion appears to be missing a key factor—public opinion is king, and I fear many will not be honest with you and call this out.
2026-11-14 · assertive framing · Politics Chat, November 13, 2026
Because jurors were drawn from voting lists, being purged from them meant Black Americans couldn’t sit on juries, either.
2026-11-13 · assertive framing · November 12, 2026
After the unfortunate results of the recent Midterm Elections, The MAGA movement took a blow, but I will not falter….
2026-11-12 · assertive framing · November 11, 2026
While the Democrats will remain in the minority in Congress until January, they are making it clear they intend to hit the ground running.
2026-11-11 · assertive framing · November 10, 2026
In another post he suggested just what that “worse” might be.
2026-11-09 · assertive framing · November 8, 2026
Also by Heather Cox Richardson
November 19, 2026
2026-11-20 · Letters from an American
Politics Chat, November 19, 2026
2026-11-20 · Letters from an American
November 17, 2026
2026-11-18 · Letters from an American
Politics Chat, November 17, 2026
2026-11-18 · Letters from an American
Nothing else under this byline is closely related to this article, so these are simply their most recent.
All 139 articles by Heather Cox Richardson →

Topics

Amodei Anthropic Face the Nation OpenAI–Hugging Face the New York Times

Subjects

Amodei PERSON · 5× Anthropic ORG · 5× Coxon ORG · 2× OpenAI ORG · 2× Dario Amodei PERSON · 1× Face the Nation ORG · 1× Jo Ling Kent PERSON · 1× Kate Brennan PERSON · 1× Mike Isaac PERSON · 1× the New York Times ORG · 1×

Narrative

By this century, the authors note, the conditions of the 1990s were reversed: baby boomers are retiring and productivity is slowing. And yet Treasury Secretary Scott Bessent told an audience at Southern Methodist University last week that he expects to see 3% growth again after the end of the war on Iran.
framing: assertive · carried by 1 article(s) · first seen 2026-09-14
He noted that July’s OpenAI–Hugging Face incident, in which programs designed to hack into systems found weaknesses that permitted them to escape the “sandbox” in which designers were testing them for about a week before anyone noticed had, luckily, been relatively harmless, but warned that “in my opinion, a swarm that possessed greater capabilities but a similar level of misalignment could have caused catastrophic damage.”
2026-09-14 · Letters from an American
September 13, 2026 · assertive framing

Claims (65 extracted, 10 hedged)

On Saturday, September 12, Dario Amodei, the chief executive officer of the artificial intelligence company Anthropic, published a 3,800-word essay calling for AI companies to slow down their improvement of AI models. asserted
companies → publish → models
Amodei expressed concern that AI models are themselves pushing advances faster than engineers can understand them. asserted
engineers → express → them
He noted that July’s OpenAI–Hugging Face incident, in which programs designed to hack into systems found weaknesses that permitted them to escape the “sandbox” in which designers were testing them for about a week before anyone noticed had, luckily, been relatively harmless, but warned that “in my opinion, a swarm that possessed greater capabilities but a similar level of misalignment could have caused catastrophic damage.” uncertain
that → note → damage
Within a year, he warned, such a swarm could take over the entire internet, causing hundreds of billions of dollars in damage. uncertain
swarm → warn → damage
In his essay, Amodei called for AI companies to commit to giving embedded third-party evaluators access to their work, rather like the regulatory supervisors in banking. asserted
companies → call → banking
He called for AI companies in democratic countries to establish safety standards and limit their rate of progress. asserted
companies → call → progress
And he called for “[t]he US and other democratic governments [to] attempt to coordinate with authoritarian governments, to the extent this is possible, while taking seriously the challenges of verifying compliance.” asserted
this → call → compliance
Today on Face the Nation, Amodei explained to host Jo Ling Kent: “My view here is it has always been very strange that this technology is being built by a private company. asserted
technology → explain → company
People ask me that question all the time. asserted
People → ask → question
Why isn’t this being built by government? asserted
this → build → government
And the strangest thing about it is, I agree with them. asserted
I → agree → them
I’m uncomfortable. asserted
I → ’m → ?
Government didn’t build this technology. asserted
Government → build → technology
This company, this technology came from the private sector, and we—Anthropic, and I would hope other companies, have done everything we can to try to have legitimate oversight mechanisms. asserted
we → come → mechanisms
“I think the government and the public need to have a stake. asserted
government → think → stake
And this is why we’ve supported regulation of the technology. asserted
we → support → technology
Regulation constrains the private companies. asserted
Regulation → constrain → companies
Regulation allows the public and its elected representatives to have a say, and limits what the private companies can do.” asserted
companies → allow → what
Asked if he would be willing to give the technology itself to the government, Amodei answered that he might be willing to give it to the right combination of governments. uncertain
he → ask → governments
“I want to be very clear about this,” he said. asserted
he → want → this
“I am concerned that one single government could abuse this technology just as easily as a single company could. uncertain
company → abuse → technology
But I think a combination of democratically elected governments—I don’t know about hand over, but some kind of oversight, some kind of joint governance. asserted
I → think → governance
Again, that would be the work of years, but I wonder if that’s the direction we need to go in.” asserted
we → wonder → years
As Mike Isaac of the New York Times noted, Amodei’s call came days after an Anthropic researcher, Jacob Coxon, resigned, posting on social media that neither OpenAI nor Anthropic was “acting responsibly. asserted
OpenAI → note → media
They are racing straight to self-improving superintelligence and gambling with our lives.” asserted
They → race → lives
They are, he said, building “superhuman systems that can hack anything, revolutionize any field overnight, and acquire real power and resources.” asserted
that → say → power
“The people building AI earnestly believe that it could kill us all by the end of the decade,” Coxon said. uncertain
Coxon → build → decade
He claimed that people at OpenAI might not yet realize what they’re constructing, while those at Anthropic do understand but believe “they are locked in a race to get there first—they believe no one else will act responsibly, so they must do it themselves, despite the risk. uncertain
they → claim → risk
Critics are skeptical, suggesting that AI leaders are hyping their product to make it appear more valuable than it really is. asserted
it → suggest → product
There is widespread concern that the massive spending on AI infrastructure like data centers and chips might not deliver the revenue and profits that would justify such spending. uncertain
that → be → spending
In June, Kate Brennan, associate director of independent research institute AI Now, told Aimee Picchi of CBS News: “The returns are not coming in, and the claims that are being made, in terms of efficiency or productivity numbers, are not netting out.” uncertain
that → tell → numbers
Others wonder if Amodei’s public concerns aren’t designed to get the government to regulate AI in such a way that it creates a framework that would make it hard for smaller companies to break into the market. asserted
companies → wonder → market
Still, Kate Conger of the New York Times noted last week that in July, more than 1,300 employees from Anthropic, OpenAI, Meta, and Google’s DeepMind—all leading AI companies—signed an open letter asking the U.S. government to regulate AI to slow down its development. asserted
employees → note → development
Conger also notes that in August, more than 100 tech companies offered their models to hospitals and infrastructure systems to enable them to guard against cyberattacks powered by AI. asserted
companies → note → AI
After Amodei’s essay appeared, Sam Altman, the chief executive officer of OpenAI; Elon Musk, who has been increasing spending on AI through his SpaceX rocket company; and Demis Hassabis, chair of Google DeepMind, all posted their support for slowing down the pace of AI improvements. asserted
who → appear → improvements
On October 30, 2023, President Joe Biden issued Executive Order 14110, calling for the “Safe, Secure, and Trustworthy Development and Use of Artificial Intelligence.” asserted
Biden → issue → Intelligence
The document said AI must be safe and secure. asserted
AI → say → ?
It called for the promotion of “responsible innovation, competition, and collaboration” and specified that AI must engage U.S. workers and advance equity and civil rights as well as privacy and civil liberties. asserted
AI → call → equity
The development of AI must protect consumers, it said, and the government must make sure the technology is deployed responsibly. asserted
technology → protect → consumers
Revoking this executive order was one of the first things Trump did on January 20, 2025. asserted
Trump → revoke → January
…and 25 more, not listed.
Give feedback · no sign-in needed

No sign-in needed for now. Your note is sent without a name, so it cannot be answered.

History Support