Jacob Coxon, a young man who previously worked at OpenAI and Anthropic, has resigned from his position citing the danger of developing artificial intelligence that could potentially harm humans. In a viral post, he urged lab researchers to consider the risks and called for Americans to stop building AI altogether. Evan Hubinger, lead in charge of "alignment" at Anthropic, expressed agreement with Coxon's concerns, stating that there is a >10% chance that AI could kill all humans within the next decade. The article presents this as a pressing issue, suggesting that experts are aware of the risks but do not have a clear plan to mitigate them.
Written by the local model on 2026-09-11,
using this article's own text rather than the other coverage of the
same event (that is the story summary below).
Story summary
The author of this article is Nellie Bowles, but she doesn't seem to be discussing any major news story about AI or technology. Instead, she mentions Jacob Coxon, a young man who left OpenAI and joined Anthropic before resigning four months later. He wrote a letter explaining his reasons for leaving the field of artificial intelligence, citing concerns about its dangers. The author appears to be using this anecdote as a way to segue into her usual TGIF column, which is a weekly roundup of news and events, rather than a serious news analysis.
Written for “Employment Discrimination Case” on 2026-09-12,
grounded in this article and the 0 other(s) covering the same event.
Why this leaning score
The model judged this article politically coded and scored it -0.85, but all 2 of its quote(s) are attributed speech - words the article quotes from someone, not the article's own narration, so the score is not published.
Written under an earlier scoring contract, which gave a paragraph
rather than checkable quotes. Re-analysing this article replaces it.
Leaning score withheld for article 7692: attributed speech only · logged 2026-09-11
TGIF on September 11 reminds me of me, every year, trying to celebrate my birthday on Adolf Hitler’s birthday.
asserted
me → remind → birthday
A terrible and ruthless commander was born that day, and also Adolf Hitler.
asserted
commander → bear → ?
Someone’s always doing something terrible in honor of him, and I’m just trying to have a tiramisu.
asserted
I → do → tiramisu
(Look out for a package of amazing Free Press stories on the 9/11 anniversary in a special edition of The Front Page later this morning.)
Quickly, before we get to the news: If you haven’t heard, we are hosting another supper club.
asserted
we → look → club
And my Bar is doing an event with author and historian Simon Sebag Montefiore on October 1.
asserted
Bar → do → October
Writing resignation letters from AI companies is the chic new thing: One young man, Jacob Coxon, left OpenAI and joined Anthropic and then resigned four months later—“I resigned from Anthropic today,” he writes—before describing the danger in developing artificial intelligence and asking people to please stop.
asserted
he → write → people
“If you are a lab researcher, I urge you to consider what the next few years will actually feel like.
asserted
years → urge → what
Do you want to kick off a superintelligent RL run without a rigorous understanding of its mind?
asserted
you → want → mind
Should you put your head down because ‘it’s happening anyway’—or take this moment to call for different conditions?”
Coxon’s post went super viral.
asserted
post → put → conditions
Even Elon Musk was shocked by the virality: “I don’t think this has ever happened for a post from a new account with almost no prior activity.”
asserted
this → shock → activity
Jacob’s argument is that everyone building AI knows that it could kill us, so Americans have to stop building AI and, I guess, hope that China and other countries just voluntarily stop too?
uncertain
China → build → AI
Or that everyone in the world pinky promises to not do anything weird?
asserted
everyone → promise → anything
Evan Hubinger, the Anthropic lead in charge of “alignment,” which means making sure AI doesn’t decide to kill us, actually agreed with our Jacob on the stakes: “Jacob is correct here—we really do earnestly believe AI could kill all humans!
uncertain
AI → mean → humans
I personally think it is >10% within the next decade.
asserted
it → think → decade
I believe Anthropic is trying its best, but we do not yet have a plan to solve alignment for superintelligence and are not clearly on track to.”
asserted
we → believe → track
I agree too!
asserted
I → agree → ?
Can someone tell us what to do about it?
asserted
someone → tell → it
Read this report on “misuse” to understand what Anthropic is combating, and what Iran and the others will use AI for as soon as they can (okay, fine, partly they used it just to plan the Supreme Leader’s funeral, which is a great time to bring in Claude).
asserted
which → read → Claude
Like all arms races, everyone has to agree to stop doing the race for it to work.
asserted
it → have → race
That is not what’s happening.
asserted
what → happen → ?
But smart young Americans have been told for a generation that we are the worst people on Earth (aside from Israelis) and should cede power to any other more ethical nation (CHY-NA, Pakistan, anywhere is better than Mom’s house).
asserted
NA → tell → house
So we do have a wide swath of intelligent people ready and willing to be subjugated.
asserted
we → have → people
They want us to put all our tools down and sing “Kumbaya” until all the peoples of Earth sing together, hand in hand, or until a better leader (Kim Jong Un?) comes to guide us.
asserted
leader → want → us
Building the thing that we need to win feels a little icky.
asserted
we → build → that
What sweet, gentle Jacob (the best camp counselor in all of Maine) can’t imagine is that genuinely malevolent nations sit outside our gates, some hoping to kill our society and subjugate us to their interests.
asserted
some → imagine → interests
I don’t mean to offend you, Jacob, I just think we’ll also need to have AI tools given this human reality.
asserted
we → mean → reality
Oh god, guys, we’ve made him cry.
asserted
him → make → ?
He’s wearing his backpack on the front now.
asserted
He → wear → front
