Jacob Coxon resigns from Anthropic over AI risk
1 Sep 10 · 16d ago · 104 articles · 7 videos · 77 posts · 8 sources · development 1 of 8
Former Anthropic and OpenAI researcher Jacob Coxon resigns, saying both companies were "gambling with our lives" by racing to build systems they can't control.
“gambling with our lives”
Jacob CoxonJacob Coxon Former Anthropic/OpenAI researcher
Dario Amodei Anthropic CEOGoogle DeepMind researcher AGI safety researcher (resigned)
Kara Swisher Tech commentatorDan Selsam OpenAI capabilities researcher
The whole story articlesvideospostscomments the bright band is this development · numbered dots are the others · click one to jump
What was reported 1 claim about this development
-
first by Anadolu Agency, 16d ago · also The Hill, Associated Press, Yahoo! Finance Canada, Reuters, Investing.com News, Channel News Asia
3 more headlines
- OpenAI pushes for mandatory safety requirements after rogue agent breakouts The Hill · 16d ago
- OpenAI calls for mandatory safety requirements The Hill · 15d ago
- OpenAI pushes for mandatory national AI safety requirements Yahoo! Finance Canada · 15d ago
On air 7 segments · click to watch, arrow keys to move
What people said 24 voices · best of 36 · verbatim
-
FWIW I don’t think that AI is an existential threat, nor on track to become one. I do however think that the leaders of the biggest AI companies, AI’s strongest boosters, and some of the folks who are most adept at using it to exploit others are actively hastening along OTHER existential threats.
-
Jacob Coxon is not alone: AI Superintelligence is an Existential Threat to Humanity Recently, Jacob Coxon, a former researcher at Anthropic and OpenAI, sent shockwaves throughout the world by stating, “The people building AI earnestly believe that it could kill us all by the end
-
E
AI Doomerism has been in the news lately, especially with the loud, viral resignation of Jacob Coxon from Anthropic. Maybe I've missed it, but I haven't seen any statements from CEOs and other execs who are more on the AI Booster side of the ledger. 🧵>>
-
K
if AI developers are truly scared AI could wipe out humanity, humanity should wipe out their companies.
-
It's so weird to me to think we're all going to die from AI but not have any explanation as to how that is going to happen beyond "well the AI will just be really smart and will think of something"
-
G
Is it already too late? https://www. comicsands.com/anthropic-ai-re searcher-coxon-warning?utm_source=mastodon&utm_medium=infeed&utm_campaign=linkprogram
-
M
On Tuesday, experienced AI researcher Jacob Coxon resigned from the AI firm Anthropic—saying that both that company and OpenAI, his previous employer, were “gambling with our lives." It’s far from the first time researchers have tried to raise the alarm about AI’s dangers.
-
🇺🇸 Anthropic caught scientists using Claude for research that could help build biological weapons. These weren’t people asking a chatbot how to make a bioweapon. The requests looked like normal science, which is why they were hard to catch. In one case, a scientist wanted help
-
D
Ahhh, so the Jacob Coxon AI-doomsday tweet was a marketing astroturf campaign. # ai # astroturf “The thread hit the media immediately. Specifically, it hit the Wall Street Journal 18 minutes before the first tweet went up.” https:// pivot-to-ai.com/2026/09/10/jac ob-coxon-how-to-run-an-ai-doomsday-media-campaign/
-
P
Concerns about the risks of artificial intelligence, and whether the companies building it are doing enough to keep it safe, have erupted across the tech world.
-
OPENAI CALLS FOR MANDATORY AI SAFETY RULES OpenAI says Congress should create capability-based federal safety requirements for the handful of labs building the most advanced AI systems, including independent testing, stronger cybersecurity and incident reporting. The company says those rules sho...
-
L
What I like about all the I-pulled-the-number-out-of-my-hindquarters predictions such as "there's a 10% that AI kills us all in the next decade" is that if it doesn't, you still get to claim your prediction was correct: "there *was* a 10% chance". Rationalists' own Nostradamus.
-
R
The danger of Ai and the tech industry is warning us directly and we are not heeding that warning.
-
If this level of risk is true, at least the AI labs will say: “I told you so”. But what are our governments doing in response to these warnings? We shouldn't stop innovation but we do need to do it in a regulated, safety first way. G7? G20? UN? Are you listening?
-
T
Yesterday we posted about Anthropic researcher Jacob Coxon saying Anthropic and OpenAI were “gambling with our lives,” as he announced his departure from the company. And today, Anthropic said it has disrupted several potential plots by scientists trying to build biological weapons. So, what are the odds of AI killing us off by the end of the…
-
AI* has a >10%** chance to kill us all *not the AI we're working on but some other AI that we'll build for sure if you keep funding the company I work at **number I made up based on a still-theoretical technology that nobody has but we all love talking about
-
D
BY VOTING FOR DEMOCRATS, WE HOLD THE AI COMPANIES ACCOUNTABLE! #ai #noAI #democracy #Democrats #vote #VoteBlue #elections #midterms #uspol #uspolitics washingtonsun.com/congress/dem... Democrats in Congress Plan to ...
-
“We really do EARNESTLY believe AI could kill all humans!” “We really do EARNESTLY plan on finalizing a pre-IPO revolving credit facility and targeting a public market debut potentially by mid-to-late October 2026!”
-
T
#AI #lawyers #legal #job www.technologyreview.com/2025/12/15/1... AI might not be coming for law...
-
Former OpenAI and Anthropic researcher Jacob Coxon, who publicly resigned from Anthropic yesterday with a stark warning about AI safety, was interviewed by Anderson Cooper on his show this evening.
-
M
I've been contacted at least twice in the past 24 hours by friends and family in disparate social circles about these recent AI news items, embodied by this Ars article. On one hand, I appreciate that I'm someone to these folks who is apparently both trustworthy with their fears and also trustworthy with this topic. On the other, I tend to give…
-
AI isn't going to wipe out humans. But AI doomers have a shot at hamstringing the most incredible advancement for humanity since the steam engine. We can't let that happen.
-
K
(paywall) I didn't think that it would happen, but I agree with Musk on this one. The recent calls for an AI slowdown seem more like an attempt to artificially create a barrier for others to catch up with OpenAI and Anthropic. # AI # regulation https://www. forbes.com/sites/siladityaray/…
-
It's a bewildering but true fact — many people working within the big AI companies are earnestly very concerned that what they are building could literally kill everyone.
All 8 developments of OpenAI, Anthropic, Google Confirm Secret AI Safety Talks… →
NewswiresMastodonXHacker NewsBlueskyGoogle NewsYouTubeReddit