"Knowledge is a single point, but the ignorant have multiplied it."
America
Joined March 2015
- Tweets307K
- Following5.4K
- Followers8.9K
- Likes141K
Pinned Tweet
Interesting. Speaking of monitoring Claude usage, you realize you and the other frontier labs could shut down China's distillation attacks by invisibly poisoning the data?
And the government might even require you to do so?
ralphcerchione.substack.com/…
Ralph Cerchione retweeted
A favorite optimistic view of the future of AI is from a pessimistic book that not about AI: "Rise and Fall of American Growth," which argues that we had a one-time explosion in living standards from 1870 and 1920 (& later outside the US).
In 1870: no toilets, no city plumbing, lamp light, no central heating, little public health, horses for local travel, etc. In 50 years, all of this was solved & the nation was electrified. In the next fifty modern medicine, air travel and telecommunication happened.
The book then points out that we have had no changes in lifestyle of similar scale since the 1970s. Written in 2016, it argued that we would likely never experience another boom like the "Special Century." The positive view of AI should make us ask "But what if we do?" and "How do we make it all go right?"
Ralph Cerchione retweeted
JUST IN: Anthropic just agreed to pay $3.4 billion for gas-fired generators to power its data centers, per the FT.
That's 1.7 gigawatts of Caterpillar equipment, roughly enough electricity for over a million homes. It's the largest known single purchase of power generation equipment by an AI lab.
AI companies aren't waiting for the grid anymore.
Ralph Cerchione retweeted
For the MSS, trying to get close to Jensen Huang would not be about trade secrets. This would be a political case: both from access to political insights in DC and for political influence.
The first is bad enough, because of the circles he ran in. The second though is what we often forget about the MSS: they want to influence as much as collect intelligence.
Nvidia sits at the crux of a core technology that Beijing has determined has near existential consequences for their view of global rivalry and underpins comprehensive national power going forward. How do they not try to influence, given how the MSS views their mission?
If the MSS tried to get close to Huang, then there are many others who should be thinking about counterintelligence risks that probably have not been.
Hallmarks of a smear campaign: anonymous sources, a bit salacious, compliant journalists, leaking side not winning the debate, vindictive, innuendo not evidence. If Huang provided information that gave China an advantage, say so.
ft.com/content/a55a2a22-4479…
Ralph Cerchione retweeted
Speaking generally: the “pattern of life,” foibles, likes and dislikes, private predilections and biases, and so on, of the ultra powerful and well-connected can be invaluable intelligence in itself. It’s not all about documents with classification markings.
Nvidia will investigate CEO Jensen Huang’s relationship with possible Chinese spy
...the company said it had “no reason to believe that either national security or Nvidia secrets were compromised.”
techcrunch.com/2026/11/08/nv…
Ralph Cerchione retweeted
The issue at hand is not whether sensitive information about Nvidia or its products might have been compromised by Chinese intelligence services (though that is obviously a concern).
The issue at hand is whether Mr. Huang's confidential personal relationship with his customer might have weighed materially on any decision to export Nvidia's products to Megaspeed—or to defend the company's broader client relationship with Megaspeed—even after it was revealed the Commerce Department was investigating that company for surreptitiously diverting Nvidia's products for resale to China, in contravention of U.S. law.
Any independent investigation should get to the bottom of that very obvious apparent conflict of interest, and spillover risk to U.S. national security.
Nvidia’s board stated today that it has found “no reason to believe that either national security or Nvidia secrets were compromised” as a result of Jensen’s actions.
They are initiating an independent investigation overseen by a committee of independent directors, assisted by outside independent counsel, that will examine all available information. This will be an unbiased, exhaustive and thorough investigation given the fiduciary responsibilities of these independent directors.
Jensen said that “I support the investigation and will cooperate fully. I love our country and our company. I am committed to protecting NVIDIA and supporting America’s national security. I am proud and grateful to be an American.”
I have known Jensen for almost 30 years. I simply do not believe he would ever knowingly do anything to disadvantage either America or Nvidia.
The only thing he has ever said to me about China is the absolute importance of keeping the American hardware ecosystem, led by Nvidia, ahead of China. That’s it.
More detailed thoughts follow below:
Ralph Cerchione retweeted
NINETY-NINE PERCENT of Australia’s international internet traffic travels through about 15 cables on the seafloor. As @RichardMarlesMP put it, they’re the arteries of modern life, and they can be cut by an anchor at night.
Protecting them is one more reason to deepen our cooperation through #AUKUS.
nine.com.au/australia-news/a…
Ralph Cerchione retweeted
Been working for this moment for over a decade. What I've witnessed behind the scenes, together with what we have been building, tells me we are about to enter a truly incredible era of AI-powered scientific discovery.
It's an incredible time to be doing research.
I will be sharing more research, thoughts, tools, workflows, and papers around scientific discovery in the coming weeks.
Across our projects @GoogleDeepMind, we are witnessing a genuine phase transition in the pace of scientific discovery. You will see a glimpse of this in our releases in the next few weeks.
Ralph Cerchione retweeted
We’re moving from an era of AI where the output comes roughly at the rate that you can chat back and forth with the AI to work with you, to a period where you can set a problem for a set of agents to go off and do an arbitrarily large amount of work for you. This means productivity gains that are maybe 20-50% to gains that don’t really have an upper limit.
Agent swarms and the ability to fan out complex work across many agents doing tasks in parallel provide a massive unlock in a lot of spaces. This will lead to qualitatively different outcomes in my fields.
This has already been seen across the coding domain. Next, cyber is almost entirely a problem space where more agents reviewing more things will lead to far better outcomes. We’re seeing the same now in many scientific fields, and this almost certainly will change life sciences.
But this will increasingly be true in any field where you can parallelize computer-based work to complete something faster or do something at a totally different quality level. Which is actually are some of the most valuable domains of knowledge work today.
Our brains aren’t really wired for the new scenario we’re heading into where compute becomes the variable that we can flex to solve any problem that we want. We’re used to thinking that the amount of time that goes into a task is correlated to the output that we get from it. And most domains of knowledge work or science have generally always been correlated with this. That’s changing forever.
Ralph Cerchione retweeted
OpenAI posted a chart for the internal model that solved Navier–Stokes (the model I call GPT-Aeon) with the NS announcement. There were no diminishing returns on test-time compute, it was still going up. I will quote from the blog:
'Since August 28 we have been training a new internal model that has exhibited unprecedented performance in our benchmarks, including mathematics. This model’s training is ongoing and its performance continues to improve.'
That was two months ago.
Ralph Cerchione retweeted
Great piece by Henry Farrell.
"So why do I think that the machine god approach, far from being an “epistemic bare minimum,” is a pernicious brain-devourer? The answer is that the metaphor shrinks rather than expands your horizons of the possible, forcing you to focus on the single overwhelming question of what Machine Jesus is going to do as soon as he descends from that gleaming cloud. Is he going to usher in paradise on earth? Or is he actually a shoggoth in a Jesus-mask, condemning us to enslavement or destruction? Once you have accepted that Machine Jesus is in the process of descending, all questions besides determining his intentions for humanity seem irrelevant."
programmablemutter.com/p/mac…
I'm writing my own piece on how a lot of metaphors and analogies are skewing the discourse, should hoepfully be out soon!
Ralph Cerchione retweeted
Good for Nvidia’s board for doing an investigation. But Jensen’s behavior is only half the story.
The biggest problem remains unaddressed: the Department of Commerce has enacted policies—at the encouragement of Jensen, and which the MSS reportedly sought as well—that allow China to continue to access the most advanced Nvidia chips, including via Megaspeed. @CommerceGov leadership knew that Jensen was sleeping with a Chinese spy and that Megaspeed was a Chinese cutout, but still actively opposed any policies to tighten export controls on Nvidia chips. Why?
Commerce could have blocked chip shipments to Megaspeed the moment it found out that it was run by a Chinese spy, but it did not. Commerce has also refused to enforce regulations—which ironically remain on the books—that would require exporters to apply for a license to ship large numbers of AI chips to untrusted entities in many countries, which would prevent China from using shell companies like Megaspeed to buy controlled chips. And Commerce has done nothing to cut off China’s remote access of controlled chips, or its ability to buy controlled chips in the United States.
Why is Commerce protecting Jensen and supporting his preferred policies even though it has known for months that Jensen was sleeping with a Chinese spy? Why is it STILL doing so?
Obviously Nvidia needs to take significant steps in its operations, security, and policy advocacy to ensure it takes national security concerns more seriously. But it is even more important that the Department of Commerce tighten up export controls to prevent things like this from ever happening again. It is utterly baffling that despite being privy to this information for months, they have not taken even the most basic actions to protect the country, such as cutting off a company that is literally run by Chinese spies from buying AI chips.
By the way, as of this writing, Nvidia STILL lists Megaspeed as a “regional cloud partner” on its website.
Breaking News: Nvidia opened an investigation into Jensen Huang, its CEO, after he acknowledged to the company’s board that he had been intimate with a business executive who U.S. intelligence determined to be a Chinese spy. nyti.ms/3ViPMAt
Ralph Cerchione retweeted
as abundant intelligence permeates into the physical world and accelerates engineering (via code), we are going to see: safer cars, less expensive homes, more durable products, etc
Ralph Cerchione retweeted
Great timing for Anthropic to launch its IPO media campaign on Monday. They were supposed to do this in September, then October, followed by weeks of reports that they planned to go public before the midterms. It just kept getting pushed back.
Ralph Cerchione retweeted
NEW: More drama on the Pentagon's 1260H list. A US judge blocked the Defense Department from enforcing its designation of WuXi AppTec as a Chinese military company, granting temporary relief to the major pharmaceutical contractor:
bloomberg.com/news/articles/… with @Suvajourno
Ralph Cerchione retweeted
We need to stop using generic terminology. AI is not a monolith. There are critical things we need to do with AI for economic prosperity, there are things we may have to do for reasons of national competition, there are things we will at some point need to do carefully for the sake of charting and taming the computational universe, and there are things we should totally avoid doing at this stage of history, and maybe ever, if at all possible.
If we can’t be precise about these issues because we’re saying vague things like “using coding agents to make AI progress should be illegal”, then we risk failing altogether. Each of the goals above depends on achieving the others.
If we can’t prosper we can’t lead the world into the future. Everyone needs to internalize this. Democrats cannot subjugate themselves to antigrowth. Republicans cannot subjugate themselves to technocapital. Both roads lead to despair.
