WSJ report: Effective altruism deeply shaped Anthropic's AI-safety culture
A Wall Street Journal investigation finds EA doomerism embedded in Anthropic's founding, with some early employees weighing remote land purchases as a hedge against AI catastrophe.
What to know
- A WSJ investigation traces how effective altruism's existential-risk fears shaped Anthropic's founding culture and safety philosophy.
- Some early Anthropic employees are reportedly considering buying remote US land to relocate to if AI development goes badly.
- The story probes the apparent contradiction of AI 'doomers' who nonetheless build the technology they fear could be catastrophic.
Anthropic AI company at center of the reportKeach Hagey Wall Street Journal reporterBerber Jin Wall Street Journal reporterEarly Anthropic employees Subjects of the report
How it unfolded 1 development · click the chart to see its coverage articlesposts
-
1
Reporter explains why doomers still build the technology they fear
WSJ reporter Keach Hagey, describing the reporting process with colleague Berber Jin, framed the story as answering why people who believe AI has a meaningful chance of killing everyone are nonetheless building it.
“if people at Anthropic believe that AI has a meaningful chance of killing us all, why are they building it?”
— @keachhagey -
first by Business Standard, 17d ago · also PBS NewsHour, The Hindu, AP News, Japan Today, Philadelphia Inquirer
1 more headline
- Anthropic and OpenAI sound the alarm on AI safety — and seek to shape how it's controlled PBS NewsHour · 16d ago
-
S
AI companies have shown themselves to be unqualified to keep track of their models - so why are we letting them be custodians of reporting incidents? My latest for @theguardian.com
2 more of the top 3 · 12 posts in this stretch
-
B
As tech titans warned an AI-weary world that their own advanced systems could endanger humanity, the question emerged: What do they have to gain? https:// broadbandbreakfast.com/anthrop ic-and-openai-sound-the-alarm-on-ai-safety-and-seek-to-shape-its-control/ # ai # artificialintelligence # politics # UnitedNations # world # news # technews #…
-
Over the last few weeks, so many people have asked me: if people at Anthropic believe that AI has a meaningful chance of killing us all, why are they building it? @berber_jin1 and I did a deep dive into Anthropic's relationship with effective altruism to try to explain how doomerism came to be so ...
-
-
background
WSJ publishes investigation into EA's influence on Anthropic — The Wall Street Journal detailed how effective altruism's fears about existential AI risk have shaped Anthropic's safety culture and internal thinking, including reports that some early employees are considering buying remote US land for relocation if AI outcomes turn catastrophic.
Also covered reported alongside — the timeline has no entry for these yet
-
3 outlets Tens of thousands of security probes show OpenAI's Hugging Face incident was just the beginning
first by The Decoder, 17d ago · also International Business Times, Tom's Hardware
2 more headlines
- OpenAI, Anthropic Face a Growing AI Security Problem. Tens of Thousands of New Incidents Under Investigation. International Business Times · 16d ago
- OpenAI and Anthropic are reportedly investigating tens of thousands of AI security incidents; OpenAI pauses testing after AI 'kill switch' fails to stop a rogue agent Tom's Hardware · 16d ago
-
first by Guardian Tech, 16d ago · also The Guardian
1 more headline
and 4 smaller pieces
What people are saying 12 voices from 4 sites · best of 16 · verbatim
- Sep 30
-
Anyone slightly qualified in CS knows that the only fully reliable kill switch is physically unplugging the damn thing from ethernet/power. Any software-based “kill switch” is fundamentally flawed.
- Sep 29
-
Ai is dying. Thats whu there shutting everything down. They dont have money anymore for it and are seeing no roi. Its dying out.
-
So OpenAI pump faked Anthropic into IPOing which has now exposed just how awful their metrics are. While OpenAI stays private for longer to farm aura off of its insane growth and reach profitability. It now looks likely Anthropic goes below 1T shortly after IPO while OpenAI jester maxes its way t...
-
Two thoughts: — 1. OpenAI is doing live tool calls during long-horizon RL training runs?!? That might explain some of the external cyber attacks. — 2. Why isn't Anthropic having these troubles? [embedded post]
- Sep 28
-
Based on my own reporting on OpenAI's audited financials from 2025, it appears that Anthropic was a worse business (at least in that year), spending $2.75 make $1 versus OpenAI spending $2.60 to make $1. Could change in 2026! https://www.wheresyoured.at/ ...
-
S
What Anthropic and OpenAI stand to gain from calls for an AI slowdown | AP News https:// apnews.com/article/ai-slowdown -midterms-anthropic-openai-ipo-9a057de94eb8f30a2fdb5b938918627e
-
OpenAI and Anthropic probe tens of thousands of AI incidents, Axios reports https:// thenextweb.com/news/openai-ant hropic-tens-of-thousands-ai-incidents-axios?utm_source=flipboard&utm_medium=activitypub Posted into TNW - All Stories @ tnw-all-stories-thenextweb
- Sep 27
-
Effective altruism and its fears that AI could wipe out humanity have wielded big influence at Anthropic.
- Sep 26
-
Our sourcing is from within the labs and from independent researchers. The tens of thousands number is referring to a totality of incidents at both Anthropic and OpenAI, not just at Anthropic as your post here indicates. Including the pause from OpenAI was relevant for a wide variety of reasons.
-
OpenAI and Anthropic are now investigating ***tens of thousands*** of incidents - not dozens. “The total could grow well beyond tens of thousands.”
-
i know Anthropic has had some issues of its own, but OpenAI's problems seem orders of magnitude worse. why?
- Sep 23
-
hot take: OpenAI could go bankrupt soon from lawsuits and hack payouts. This just shows the real gap between Anthropic and OpenAI. The talent gap is huge, but their alignment gap is even bigger.