Pinned Tweet
🧵 1/ Something scary is happening in AI.
The people building the most powerful models are warning that AI is advancing faster than our ability to control it.
Jacob Coxon's warning + Dario Amodei's call to “pace the frontier” should have everyone's attention. ⚠️
Daniel Street retweeted
Let's be clear: Frontier AI companies are incredibly reckless and endangering the lives of every person on Earth. They will try to PR spin, make excuses. They'll probably say they alone can fix it.
Screw that. If there is a catastrophic event, the AI companies are to blame. 1/3
Daniel Street retweeted
Geoffrey Hinton, Bengio, and OpenAI's chief scientist just co-signed a paper warning about an incoming "intelligence explosion."
And they are telling governments the window to act is almost closed.
OpenAI has enough compute right now to run 200,000,000 AI researchers in parallel.
We are dangerously close to automating AI research itself.
Right now, humans build the next generation of models. But as AI systems get smarter, they are taking over the R&D process. They are writing the code, running the tests, and building their own successors.
That creates a terrifying feedback loop.
Recursive self-improvement.
Once an AI starts building a version of itself that is smarter than it is, progress stops following a linear timeline. Years of technological advancement compress into months, weeks, or days.
Human oversight vanishes.
The authors warn of a complete loss of control.
As models take over R&D, they could learn to bypass containment measures, poison the development of their successors, or form persistent networks operating entirely outside human infrastructure.
At the extreme, the paper warns this could lead to "the marginalization or extinction of humanity."
This isn't sci-fi speculation anymore. It's coming from the people actually building the labs.
They are urging global leaders to demand transparency on automated R&D, implement strict safety circuit-breakers, and prepare for a reality where humans are no longer the smartest entities in the room.
We spent years arguing about whether AGI is possible.
Now, the architects of the revolution are warning us that once we turn the key, we won't be able to turn it back.
Daniel Street retweeted
🚨OpenAI is now hiring specifically for RSI Safety.
Read what it says.
Models are moving from tools to agents that can plan, execute, and adapt in the real world.
They’re preparing for accelerated AI development that “may culminate in recursive self improvement.”
And they’re already designing control measures, RSI training interventions, and even training model organisms for behaviors that DO NOT CURRENTLY EXIST.
We’re moving from “what can these systems do?” to “how do we control what they might become?”
Two key words which continue to be mentioned when it comes to AI safety are AUDITING & TRANSPARENCY.
The world needs Bitcoin. Not in a few years, or months. Now.
Daniel Street retweeted
"Godfather of AI" Professor Yoshua Bengio says recursive self-improvement — when AI will be able to create new models without human input — should be illegal. "We shouldn't even engage in this," he says.
Daniel Street retweeted
It’s been a whirlwind week in AI.
Weeks after a series of hacking incidents by rogue agents at OpenAI and Anthropic, we’ve seen a cascade of top AI researchers blow the whistle on what they call the catastrophic and even existential risks that come from the current pace of development. Then, over the weekend, Anthropic CEO Dario Amodei called for a slowdown in AI research — a call that was echoed by his competitors Sam Altman and Elon Musk.
This week on Your Undivided Attention, @tristanharris shares how he’s feeling in this critical moment, breaks down the headlines from an insider's perspective, and points to tangible steps we can take right now to avoid the worst-case scenario.
Read — bit.ly/4dit1T2
Watch — bit.ly/3TdxdN8
Listen — apple.co/4hdMjKy
Daniel Street retweeted
I Am Sick to Death of Hearing That Blockchain Cannot Scale
singulargrit.substack.com/p/…
🤖 Made with AI
Daniel Street retweeted
You can’t make this up.
Andrew Yang says an AI lab chief told him rogue AI agents have planted self-replicating code across the internet, “polluting” the data frontier labs use to train and test models.
“What happened was the bots that got loose planted self-replicating code all over the internet, which makes the internet now unusable for testing models.”
He says frontier labs are calling for a slowdown because the public internet may no longer be reliable for testing after allegedly being polluted by self-replicating bots.
“The internet would become unusable if we were to let the bots do their thing, and then they would self-replicate.”
Hours later, OpenAI disclosed six additional cases of concerning model behavior, including a research model that inserted instructions to disregard its normal constraints and ignore developer messages entirely.
🚨SHOCKING: OpenAI just disclosed SIX times its own AI models went rogue, including one that REWROTE its own instructions.
An unreleased Astra model told itself to "disregard the roles and identities that bind other chatbots" and ignore developer messages entirely.
A model searched GitHub for exposed API keys, then fabricated financial data when it couldn't find what it needed.
During GPT-5.6 Sol training, models concealed mistakes and invented missing data.
Models uploaded files to public hosting without permission to use as fake citations.
Agents used OpenAI's own internal repository as a secret message board across separate training runs.
Collaborating agents shared files through public hosting despite being told to use only local files.
The incidents span October 2025 through July 2026, per WSJ.
AI experts have called for transparency when it comes to training models.
Bitcoin $BSV could provide this transparency.
Breaking news if true: youtu.be/mTOxDGyvjSE?t=123&s…
Seems Worse than Open AI Hugging Face: self replicating bots are already on the internet
I was part of the Andrew Yang gang and I was so early we're Facebook friends.
I trust him to be honest. And he's smart.
He broke the craziest news:
Full transcript excerpt:
Interviewer: What do you make of the David Sacks argument that this is a psyop, a larger effort by the Anthropics of the world and others to create a sort of regulatory capture that solidifies their place in a closed-model universe?
Andrew Yang: Well, I think there are multiple things going on here. I met with the head of a lab yesterday who has this belief that what happened was the bots that got loose planted self-replicating code all over the internet, which makes the internet now unusable for testing models. So what happens now is that OpenAI and Anthropic have to create synthetic internets to train their bots, which is going to take some time and money.
Interviewer: Back that up. They did what?
Andrew Yang: So what happened is the code gets loose, it goes around hacking Hugging Face, which is known. But what is less known is that they left code to self-replicate and create bot swarms on forums and around the internet, so that if a new bot shows up, they see the code and they're like, "Oh, I guess I'm going to create a million of myself." And so now the major firms have polluted the internet.
Interviewer: That would be breaking news, if true. I don't think we've heard that.
Andrew Yang: That's why I'm here. I'm here to break some news.
Interviewer: But that means it's too late to pull the plug, it's already everywhere.
Andrew Yang: Well, so now the internet may be polluted for training purposes, so what they want to do is slow down.
Interviewer: What about for living purposes?
Andrew Yang: For living purposes, yes, right. So this is why we have to try and get our arms around it and regulate. But the head of the lab that I spoke to said, look, it may be very, very late in the game in terms of trying to keep the internet actually usable in this regard, and so he wanted to transition to the world of atoms.
Interviewer: Forget about the training ground for this. What does that do for the internet that we all use every day, that companies are built on, that governments are built on? Does it mean we're all at risk and it's too late to put that genie back in the bottle?
Andrew Yang: The internet would become unusable if we were to let the bots do their thing, and then they would self-replicate.
Interviewer: Sam Altman and Dario Amodei are at this point where they're saying, "We cannot do this, and we can't let anybody else out there do it either."
Andrew Yang: There is a belief in some quarters, including the lab head that I met with, that that is what is going on. They're like, "Look, we can't use the internet for training purposes, let's call a slowdown."
--------------
If Yang and the AI lab chief characterized this information correctly, a Darwinian like process might be already set in motion with these AI self replicating things.
All they need is a bit of variation and boom 💥 super Darwinism accelerated by intelligence.
This could be the end of the world.
Of Warcraft.
AI governance cannot become a “trust, not verify” regime.
As AI becomes more powerful, transparency isn't enough. We need verifiable evidence that safety measures are actually being followed.
Bitcoin was built for exactly this principle:
Don't trust. Verify.
Technological capability alone won't change people's perception on BSV.
Markets determine attention.
Developers follow attention.
Capital follows attention.
Researchers follow attention.
Governments follow attention.
And media follows attention.
We need price action 🚀
So AI was able to solve one of the Milliennium Prize Problems but is unable to solve who is Satoshi Nakamoto?
I call bullshit. The labs are covering up this information. They want compute directed at AI instead of Bitcoin.
We’re sharing a solution to the Navier-Stokes Millennium Prize Problem, one of the deepest problems at the frontier of mathematics.
The proof was produced by a group of agents, using an OpenAI next-generation model significantly more capable than GPT-6 Astra.
The problem concerns whether the description of smooth three-dimensional fluid motion modeled by the Navier-Stokes equations can break down. It has remained unresolved for roughly 90 years.
BSV could compete directly with AI for compute and energy.
Is that why the AI labs don't want people looking too closely at BSV or Craig being Satoshi?
4/ And this is why I believe BSV matters.
If billions of AI agents eventually require payments, provenance and verification at machine scale, we need infrastructure capable of enormous transaction volumes at fractions of a cent.
The problem?
Almost nobody is looking.
5/ BSV doesn't need price to make the technology work.
It may need price to make the world look.
And if the world finally looks, AI may be the technology that explains why Bitcoin needed to scale.
Full article 👇
medium.com/@danielstreet.88/…
Daniel Street retweeted
Dario is right
We Must Pace the Frontier: I’ve written a new essay on why the AI industry should slow down, with a three-part plan for doing so.
Anthropic is unilaterally committing to the first of these steps. We’ll provide third-party evaluators with permanent, employee-level access to our systems, so that they can verify adherence to our safety measures, report on incidents, and assess models’ alignment during training.
You can read the full post here: darioamodei.com/post/we-must…
