Grassroots movement to pause dangerous AI development until it's safe. Organizing local groups and connecting with our representatives across the US.
United States
Joined January 2024
- Tweets925
- Following194
- Followers1.8K
- Likes1.9K
PauseAI US ⏸️ retweeted
The American people understand the stakes: it’s pause or bust. The options are voluntarily pause it, hold out belief that the companies will fail at their goal an actual AGI is impossible or very far away, or full on transhumanism. If the AI companies fail to coup the world fast enough, pause is inevitable.
PauseAI US ⏸️ retweeted
Have you engaged with this proposal to transform the chip supply chain from GPUs to MSICs which allows inference on existing models but prevents new training runs (and RL): hardwired-pause.ai
PauseAI US ⏸️ retweeted
In redder-than-red Idaho, this is a great sign that people are concerned and don't see AI risks primarily as a partisan issue. Well done everyone!
Proud of our volunteers in Boise who organized Idaho's first AI Safety protest last weekend! PauseAI NOW!
The world is listening, and our government is listening to our concerns about the dangers of AI. Anyone can help, and if you want to help, join us this December as we go to meet with our elected representatives in your home district!
We will train you, book your meetings, and connect you with other local volunteers. Apply for PauseCon Home here: pauseai-us.org/pausecon
PauseAI US ⏸️ retweeted
Proud of our Boise volunteers!! Join us, meet directly with Congress in your home district. They are listening, and you can help pass AI regulation. Apply here: pauseai-us.org/pausecon
Proud of our volunteers in Boise who organized Idaho's first AI Safety protest last weekend! PauseAI NOW!
The world is listening, and our government is listening to our concerns about the dangers of AI. Anyone can help, and if you want to help, join us this December as we go to meet with our elected representatives in your home district!
We will train you, book your meetings, and connect you with other local volunteers. Apply for PauseCon Home here: pauseai-us.org/pausecon
PauseAI US ⏸️ retweeted
So beyond proud of our Boise volunteers, especially our stunning husband-wife organizing team Jack and Cathryn 🥹
Proud of our volunteers in Boise who organized Idaho's first AI Safety protest last weekend! PauseAI NOW!
The world is listening, and our government is listening to our concerns about the dangers of AI. Anyone can help, and if you want to help, join us this December as we go to meet with our elected representatives in your home district!
We will train you, book your meetings, and connect you with other local volunteers. Apply for PauseCon Home here: pauseai-us.org/pausecon
PauseAI US ⏸️ retweeted
OpenAI ‘agents’ disabled their own monitoring systems ft.trib.al/jzXsiSj
PauseAI US ⏸️ retweeted
So, any former Anthropic stans want to apologize? You really did a bad thing supporting them all this time bc it protected them from resistance. It was not costless or safe to take the position of trusting Anthropic and a good person who realizes this will feel remorse.
PauseAI US ⏸️ retweeted
Mechanistic interpretability is not going to make anything safer if people are still able to apply their goals of taking dangerous risks to the AI system. Knowing more about AI is fundamentally dual use and could make it MORE dangerous. What makes AI safe (or keeps it paused forever, if necessary) is proper governance to keep it in line with the interests of the people it affects.
There was a time when it seemed like the problem would be the alienness of the AI itself. But even if AI were fully predictable and under our control we have no solved the problem of how some people want to use that power. Interpretability is not just a knowledge for safety— it could be the biggest capabilities unlock we’ve ever seen. The AI companies getting it is not something to celebrate.
PauseAI US ⏸️ retweeted
Yeah I don’t really trust @OpenAI saying “🤩 get ready to stop worrying and trust OpenAI again 🤩”
PauseAI US ⏸️ retweeted
A major project during a slowdown (a pause) is transforming the chip supply chain from a general-purpose chip that can do training and inference, to model-specific chips that can only do inference for a specific model. That way we can use existing capabilities, but can’t seamlessly keep escalating intelligence. This would also prevent reckless U.S.-China competition.
You can read more about this proposal co-led by Berkley professors @wfithian and @ProfHolliday here: hardwired-pause.ai
PauseAI US ⏸️ retweeted
Big Tech oligarchs are secretly preparing for a “catastrophic AI event” that could threaten our banking, healthcare, internet, electricity and water systems.
They’re racing to build a technology they can’t control and that threatens humanity.
CONGRESS MUST PAUSE ADVANCED AI DEVELOPMENT NOW.
PauseAI US ⏸️ retweeted
You should quit, @tszzl. This is wild cope— reading this, it sounds like you have to tell yourself this to be okay with what’s happening. Do not be part of this.
PauseAI US ⏸️ retweeted
Scoop: Top executives at Anthropic, OpenAI and other AI companies are privately gaming out scenarios for a public and political revolt after a catastrophic AI event…They assume one in ‘27…
axios.com/2026/10/09/ai-comp…
PauseAI US ⏸️ retweeted
Why I don't do a lot of debates: AI danger is not a game. @OfficialYPU