"Knowledge is a single point, but the ignorant have multiplied it."
America
Joined March 2015
- Tweets307K
- Following5.4K
- Followers8.9K
- Likes141K
Ralph Cerchione retweeted
I'm Sam, one of Anthropic's co-founders, and I lead much of our safety work. I'm going to have more to share on that work over the coming months.
I thought I'd start by sharing a few thoughts on the latest report from @METR_Evals.
The speed of progress in model capabilities is one of the most important questions in AI safety, and METR’s work here provides the public with an independent, external view into how this is happening inside Anthropic. I’ll talk more about some of the details below, but first some context for those new to this work.
How quickly AI progress is accelerating has many important safety implications, since it determines how quickly the risks from advanced models increase. But, because most research happens within the frontier labs, it’s difficult for external observers and policy makers to get a strong sense of what that rate is. We kicked off the process with METR that led to this report earlier in the year, in anticipation of progress continuing to accelerate.
For this report, we gave METR access to a range of sensitive internal data, including benchmark scores for unreleased models, estimates of how models are progressing, and interviews with our researchers. They had complete independence to form their own conclusions.
Many of their conclusions match my own. AI is speeding up our research, but it’s hard to say exactly how much. While they conclude AI can't yet make the most important contributions to model training on its own, they think the pace of development could, by default, accelerate further.
This work began well before and is separate from the embedded third-party evaluators that are part of our commitment to pacing the frontier, but it’s a good illustration of why we think evaluators are a good idea. The more access reviewers have, the clearer a picture they can form of the situation and the better-positioned they are to highlight situations like this one.
We redacted a lot, and I accept that this makes it harder for others to check METR’s work. But we determined that sharing exactly what is driving the speedup isn’t the right thing to do, both because it’s commercially sensitive and because it risks handing out a roadmap for others to go faster. That would make the risks METR is worried about worse. That’s a tradeoff. I think we made the right call, though we’re continuing to assess the pros and cons here.
In the coming weeks, we plan to share a report with our own analysis of the speed of AI progress within Anthropic, including where we stand against our Responsible Scaling Policy, plus new steps we’re taking to strengthen our safety and security practices as AI keeps improving quickly.
Ralph Cerchione retweeted
we are incredibly, improbably lucky that our senescent authoritarian’s primary obsession is interior design
Ralph Cerchione retweeted
On Oct. 21, the team successfully locked onto a quantum signal and recorded data in flight, taking a major first step toward incorporating airborne systems in nationwide quantum networks. (3/3)
ALT Two researchers in an operations control room analyzing satellite imagery and weather tracking data displayed on large wall-mounted monitors, with one standing and pointing at the screen while the other sits at a computer workstation. Credit: NASA
Ralph Cerchione retweeted
The researchers are using NASA Glenn facilities and the PC-12 aircraft to test cutting-edge technology that could pave the way for expanding quantum networks, which could send quantum information — as well as classical data and communications — using secure, unbreakable encryption. (2/3)
ALT A person in a tan flight suit inspects the propellers of NASA's PC-12 aircraft on a paved tarmac. A person wearing a neon safety vest picks up orange cones in the background. Credit: NASA
ALT A large white faceted geodesic radome mounted on an elevated steel platform with an access ramp on a facility rooftop against a clear blue sky. Credit: NASA
Ralph Cerchione retweeted
We’re taking quantum communications to new heights! ✈️
Late last month, @NASA researchers made history when they successfully sent tiny light particles, or photons, between an aircraft and a ground station over a quantum channel, the first time this has been achieved in the U.S. (1/3)
ALT Side view of a white and blue NASA aircraft (tail number N606NA) on a paved runway taxiway under a clear sky. Credit: NASA
ALT Three NASA team members inside a communications operations room monitoring real-time flight tracking software, maps, and system telemetry across multiple computer screens and rack-mounted server hardware next to an open exterior doorway. Credit: NASA
ALT Optical tracking and laser communication equipment set up on an outdoor rooftop platform under a clear blue sky, and a blue hydraulic scissor lift cart supporting additional sensors and electronic control hardware. Credit: NASA
Ralph Cerchione retweeted
OpenAI’s Greg Brockman says "death becoming a choice" is possible, and predicts AI will make real strides on curing diseases in the next 12 to 18 months
“I think that you look at something like biology, and that’s not something that’s unique to just humans. I think that we all suffer, that the question of even just thinking about death becoming a choice. All of this is possible.”
“Having cures for diseases is something that’s become also a trope in this field, but I think that you’re gonna see real delivery and real significant strides and impacts on biology, health, and curing of diseases over the next 12 months, 18 months.”
“I think it’s gonna take time to really deliver on the full impacts, just by the nature of this field. But I think it’s all possible and doable, and it’s exactly the kind of thing we have these models that are solving these unsolved problems. Clearly, we need to apply it to biology and life sciences.”
OpenAI’s Greg Brockman reveals an unannounced JPMorgan Chase deal, with the bank using Codex and OpenAI’s cyber models to secure itself
“Well, so within the company, we were probably, I don’t know. Like, I think our enterprise revenue was about 30% versus consumer at 70%. And now we’re probably somewhere around, like, we’re majority enterprise. And it’s not that consumer has shrank. Right? It’s been growing, but it’s the enterprise segment has really taken off."
"And so I don’t know exactly how to quantify in the broader market, but you can see the way that we’re winning deals. So for example, actually, one that’s hot off the press, has not been previously announced, is that we just papered a partnership with JPMorgan Chase, and that they’re going to be using Codex."
"They’re going to be using our cyber models to secure themselves, deploying our models for enterprise transformation knowledge work across the company."
"And that for me is like such an iconic institution in such an important domain. Right? A highly regulated enterprise that trust is, like, so key to what they do and that they chose us and that we’re here to be a great partner.”
Ralph Cerchione retweeted
When an object spins in zero gravity about its middle axis, the rotation is inherently unstable, and the object periodically flips 180 degrees mid-flight, a phenomenon known as the Dzhanibekov effect.
Ralph Cerchione retweeted
You want a pause on advanced AI, regulation of the industry, and a ban on superintelligence.
China just read this post and started celebrating.
Every day America pauses, they build. Every regulation you pass, they skip. Every ban you propose, they treat as a head start.
You are not protecting the country. You are handing the most powerful technology in human history to the one nation you cannot regulate.
Ralph Cerchione retweeted
agents are going to cause a breakdown of human behavior assumptions underlying today's systems. i've started thinking about it as friction collapse. without friction, a number of norms will collapse.
consider selling tickets to a concert/event. today, you can post them at a certain time and it's first come first serve. soon everyone's agent will persistently monitor, and as soon as they are posted for sale, thousands of people will instantaneously be trying to buy the same ticket. how do you even resolve this? sell the ticket to the person whose agent gets there fractions of a second before?
we're seeing the same thing with job applications, restaurant reservations, but basically anything that relies on participation requiring effort is going to break down. sales emails/calls, participation in government benefits, apartment rentals, doctors appointments, RFP responses. customer support won't be able to handle the user who will not drop their point of dissatisfaction. with no friction, these systems just don't work anymore.
business models will break down:
- flights/hotels will be consistently cancelled and rebooked at the lowest price (great use case for your agent if you don't have this already set up)
- subscriptions will be cancelled
- retailers won't be able to arbitrage price differences as the agent will find the cheapest place to buy anything
- insurance denials (health/auto/home/etc.) will be fought tooth and nail
- bank interest rates will be optimized and avoidable charges will be avoided
a lot of what we perceive as stable social coordination is actually an equilibrium sustained by human inefficiency. rationing is a product of human slowness. signaling effort/interest/intent is a key part of our social infrastructure. inertia is a key to many business models. information asymmetry is everywhere. and attention is monetized thoroughly.
with agents, none of these things are true. humans will simply demand what they want and have their agents get it for them. there is going to be a fundamental renegotiation of how markets allocate resources, how people signal intent, and how businesses capture value.
Ralph Cerchione retweeted
The @FBI and our partners are seeing malicious cyber activity targeting industrial refrigeration and cold-chain monitoring systems around the world. These systems help keep food, medicine, and other products at safe temperatures and are widely used across critical infrastructure sectors.
Attackers are targeting electronic temperature and refrigeration controllers, particularly in grocery stores and cold storage facilities. In at least one recent incident, attackers targeted refrigeration components at three foreign blood service facilities.
While this attack was unsuccessful due to the proper alert mechanisms in place, security weaknesses in unpatched or internet-accessible controllers can allow attackers to change system settings, potentially causing inventory loss and equipment damage.
The FBI urges organizations using these systems to take the following steps:
➡️ Take controllers off the public internet
➡️ Put access behind a firewall
➡️ Change default passwords
➡️ Learn how vendors configure devices
➡️ Set safe temperature limits
➡️ Ensure alerts reach the right contacts
For technical details and more information on how to protect these systems, read our advisory: ic3.gov/CSA/2026/261105.pdf
Ralph Cerchione retweeted
Judging by the posts on my timeline, today is the big update on both RSI and ASI for many people in this community. I went through this in November 2022, so I know exactly how you feel. I know nobody drinks anymore, but I would suggest having a drink.
Ralph Cerchione retweeted
I find the “patch the planet” initiative quite inspiring. openai and others are spending huge sums of money on updating and securing all modern open source software infrastructure. it will be a continual effort to make the world more robust
This is the "upcoming cyber model" from yesterday. Besides 948 Linux vulns, it found 111 renderer RCEs in Firefox, among which, 81 were novel. 30 were exploits for bugs known to developers that were thought to be unexploitable.
We are using this technology to scan more OSS projects and secure the Internet infrastructure.
This is just a start.
Ralph Cerchione retweeted
Planning for post-quantum cryptography (PQC)? Use our new guidance + quick-reference list w/@CyberGovAu & @NSA to evaluate your vendors’ awareness of quantum-related risks, transparency & support for a secure/cost-effective/timely PQC transition 👉 go.dhs.gov/5j9
ALT Digital illustration of a pixelated key over a circuit-like blue background. The logos of the Australian Signals Directorate, Commonwealth of Australia, CISA and NSA are displayed at the top. Text reads: "Guidance: Post-Quantum Questions to Ask Your Vendors."
Ralph Cerchione retweeted
I genuinely don’t think Democrats calling for a ban on ASI understand what they’re playing with here.
Imagine China develops superintelligence and deploys millions of humanoid robots while the US voluntarily stops developing the technology.
China could eventually manufacture almost anything at a fraction of today’s cost, with factories running 24/7, AI designing better products, and robots doing practically all the physical labor.
How exactly are Western companies supposed to compete with that? Higher wages? Better productivity? Good luck.
Entire industries could collapse because they simply couldn’t compete on price anymore. We’d become increasingly dependent on China for everything from consumer goods to critical infrastructure.
And that’s before we even get into the implications for scientific research, medicine, and national security.
You don’t prevent a technological monopoly by handing your biggest geopolitical rival a head start.
That’s not AI safety. That’s economic suicide.
Ralph Cerchione retweeted
Researchers trained an AI on 680,000 brain signals..
And it figured out how to wake people from comas.
Researchers published a paper in Nature Neuroscience that crosses a terrifying and incredible threshold.
They wanted to understand how consciousness breaks down in the human brain.
So they built an adversarial AI.
They didn't give it any rules. They didn't pre-specify what it was looking for. They just fed it a massive dataset of neuroelectrophysiology and let the AI fight itself to find the hidden patterns.
What happened next is wild.
AI didn't just map the brain. It reverse-engineered the exact mechanisms behind impaired consciousness.
Then, it took it a step further.
Without a single line of code telling it to, the AI discovered an entirely new, unprogrammed treatment for disorders of consciousness.
It found a way to wake the brain back up.
We spent decades trying to manually map the mechanics of human awareness. We couldn't do it. The system was simply too complex.
So we built an AI that could.
Researchers are calling this a new framework for causal inference and therapeutic discovery. But the implications go way beyond medicine.
