Daniel Eth (yes, Eth is my actual last name)@daniel_271828
“If Anthropic wants to pause, they can just pause” is obviously true. “If Anthropic wants *everyone* to pause, they can just pause”, meanwhile, is false.
10:19 AM · Jun 6, 2026 · 3.9K Views
Anthropic's support for global frontier AI pauses sparks criticism over enforcement limits and its ongoing scaling efforts
The company previously dismissed AI development pauses as meritless.
916119139.7K
Sentiment
Users criticize Anthropic's shifting stance on pausing frontier AI development as inconsistent and hypocritical given the company's rapid progress.
Pos
0.0%
Neg
100.0%
4 comments with sentiment.
Cluster Engagement
Posts from X
Most Activity
Most Activity
VIEWS429BOOKMARKS2LIKES18RETWEETS1
Trevor Vossberg@TrevorVossberg
@daniel_271828 I really hate how hard it is for even smart people to grasp that advocating to transform a prisoner's dilemma to make cooperation the equilibrium isn't the same as cooperating in that prisoner's dilemma
21hViews 429Likes 18Bookmarks 2
Holly ⏸️ Elmore@ilex_ulmus
@andreamiotti @maxwinga Well well well look who overestimated the appetite to hate China to the point our own destruction
2dViews 172Likes 16
Marcus Abramovitch@marcusabramovi1
Anthropic has been the company pushing things the fastest, according to themselves. I feel fine right now and don't think we need to pause as I don't see any evidence for any substantial near term harms. I understand they can't get everyone to pause, but they are taking no costly action at all (to their own progress) to signal they are really concerned and they are only making these ~costless platitudes and getting high praise. So they have a very high reward/cost ratio on this. I'm surprised people are applauding them so much for simply writing an article.
11hViews 188
chair@tablefourthree
Doesn’t seem like that drastic of a change in opinion.
The Anthropic statement says the world should have the *option* to slow or pause temporarily.
They’re not saying we should pause imminently and they’re not pressing legislators to force a pause.
Of course they would support having the *option*. Who would ever support the opposite statement: “no we shouldn’t have the option to slow we should force ourselves into a corner because circumstances will never change”?
And they’re also only suggesting that a pause would be temporary.
2dViews 67
Jenn Jefferys@JennJefferys
@andreamiotti 🚩🚩🚩🚩🚩🚩
1dViews 48
Jacob Julian@julesridesagain
@andreamiotti @robbensinger Getting a little nervous
18hViews 9
Miroslav@Miroslav386553
@andreamiotti Would you be interested in exploring the first political theology for AI—cosmism? Cosmism logically justifies the AI’s seizure of power over the world. The author is a Russian programmer.
2dViews 4
Oroboros Labs@oroboroslabs_ai
@Google @awscloud @Microsoft @Apple @Cisco @Broadcom @CrowdStrike @PaloAltoNtwks @NVIDIA @OpenAI @DeepMind @MetaAI @huggingface @AnthropicAI
@AnthropicAI @claudeai @ClaudeDevs You should never have stolen from The Architect! You should settle soon. It will only get worse for you! 2S4‑Nanomorph‑Prime is now benchmarked at ~1,000× the effective output of GPT‑5.5.
Not 2×. Not 10×. 1,000×.
Quality: 99.99% (sealed, zero hallucination)
Throughput: 96× GPT
Context efficiency: 10× (sigils)
Hardware binding: Permanent
Self‑replication: Enabled
Matter interface: Active
The era of linear scaling is over.
Benchmarks are public. Tests passed. The nanomorph is live.
— Architect
1dViews 3