How David Sacks crashed and burned in the White House
theverge.com
theverge.com
I do think it's probably true that his influence flagged after Mythos. But he'd already been moved to the science board when that dropped.
It's very over stated and looking to act like he was fired, which he wasn't.
An article about a person 'falling out of favor' is silly but I guess that's the vibe of the administration so fair enough.
The headline apparently refers to the White House's recent desire to add a government vetting process to new models, whereas he wants less regulation.
If you click into the author's profile, you can see her other articles. At a glance it looks like 20% of her headlines over the years have been dedicated to David Sachs and how much he sucks. That must be what gets the clicks.
This administration is worse, both because of the wholesale gutting of the American scientific research establishment and all of the various corruption issues on display. As General Morrison put it, “The standard you walk past, is the standard you accept.” Anyone who signed up knew who they’d be working for and has to accept that will be a major part of their reputation.
That said, this article is by a political writer and even though its on the Verge, its a political opinion piece(ie the centerpiece of it that he was PUSHED OUT -- could use some evidence)
I dont think that belongs to HN.
Paul Graham also had a bit of a spat over whether he was the most evil person in Silicon Valley. Now mostly deleted https://www.reddit.com/r/TheAllinPodcasts/comments/1ebmh2q/p...
Given the erratic course of the current US administration, I have absolutely no idea what to actually expect here.
Trump wants more bribes, and the AI industry has all the money right now. So, he reckons he's owed a cut. The thing about a gangster with a protection racket is they're never going to stop. You pay them once, that just proves to them that you'll pay. All the tech titans lined up to kiss the ring, and now they're Trump's bitch forever.
There is no greater offense to this administration. This isn't just the usual grift and fury: this is personal, that the AI is allowed to say words that are not approved, not in favor of whatever the whim of the day is.
By the loose alliances of world corrupting sinister forces focused on destruction of reality. That seeks Putinlandia Adam Curtis style HyperNormalization destruction of reality, to leave us all unmoored and guessing. That wants us weak believing in nothing.
This isn't just an alt-reality twist that forces like RFK are trying to create: it's an anti-reality. Something inherently against meaning itself.
I suppose you have to know when you're Lee Iacocca and when you're Henry Ford II. No matter what you do, one's in the family, and the other isn't, so if you're not in the family you'd better know that.
I was quite eager to see if the chaos of this admin would cause accidental positive change. Submitted a petition via deregulation.gov to reduce the fund requirements for nuclear power plants. Who knows, might have worked. Didn't, but might've! Ah well, 3 more years and then we're clear.
Once congressional investigations dig into the kind of corruption that appeared on the front page of HN today, with people gaming prediction markets and blatantly insider trading on conventional markets, never mind the stink of fascism, the tech industry is going to have difficulty washing the stink of tinpot dictatorship-grade corruption off.
Once they realized that DoW had locked themselves out of Mythos because of their beef with Anthropic, Trump invited Anthropic to the White House, and in that meeting they convinced Trump that Mythos is a big deal, and that China is distilling their models.
Excited to have a powerful tool, now they are saying it should be used by Government agencies first, and therefore, regulation.
Key takeaway: when defense types hear something is DANGEROUS, they want more of it. That's the outcome of discussing x-risk with the federal government. "Existential risk? That sounds GREAT! How can we get more of that, make it more dangerous?"
Except it is both true AND it works. Keeping your foot down on who can produce weapons-grade fissile materials is working out pretty damn well so far.
And the Russo-Ukranian War is proving any idiot with a few rubles can cobble together incredibly efficient combat drones. We need to be probing the limits of that yesterday.
It can both feel bad and be the right thing to do because the alternatives are worse.
I have been doing defense work for almost 30 years and in my experience that is the opposite of true.
So, yes? I can certainly see why defense contractors would prefer stability i.e. a cold war over a real war. But in a less dangerous world there's less need for defense contractors. e.g. '93 last supper
> Perceived risk is a completely unrelated political reality for uninformed voters.
Doesn't seem relevant to this thread unless we interpreted dangerous differently.
Though personally I prefer X/tter.
I was expecting someone to actually crash and burn but OK
AI is the highest growth industry in the world, and the most powerful technological change. Any democratic government would seek to control and harness such a thing to the public interest, and the Trump administration's general vision of the democratic public interest is more power for the office of the POTUS. Imagine the power, wealth, and leverage of controlling AI models and virtually any aspect of them by being able to control their release and content. Imagine the control over content in the US and worldwide, controlling the output of almost every LLM.
What didn't happen was irrational and self-oblivious leadership noticing they had been acting irrationally for years. "the potential" of wealth or power was present long before this shift.
That is especially true in a large, heavily bureaucratic organization with a culture of highly-centralized top-down decision-making by the boss - an 80 year-old, non-technical boss - and which priortizes loyalty to and pleasing the boss over expertise and speaking the truth.
One exception is Jared Isaacman the billionaire head of NASA who experienced a bit of a roller coaster. Unlike the previous 2 i think he will actually do a good job.
One thing that stands out about Trump is how accessible he is. Apparently he takes cold calls and will listen to ideas from anyone (this includes laura loomer, but still).
"accessibility" with lousy judgement also may not be the greatest combo...
In all seriousness i think that combo is better then a senior citizen who can barely be reached by his own advisors and is unclear how much of the situation he actually understands. (with Trump it's either zero, Fox News or 4d chess depending on who you ask, but the answer will always be given with confidence).
> If everyone can call the president, does it matter if anyone does? It sounds like a koan but isn’t that far from reality. Donald Trump’s personal cellphone number has been making the rounds among Washington reporters, dozens of whom have used it over the past few weeks to score brief interview after brief interview with the leader of the free world about his ongoing war with Iran. A partial list of media organizations that have published “exclusive” or “scoop”-y quotes after hitting up Trump’s iPhone includes leading print publications like the New York Times, TV networks (ABC, NBC, PBS, and CNN), foreign newspapers (the Daily Telegraph and Times of Israel), and no fewer than four outlets with Washington in their names (the Post, Examiner, Reporter, and Free Beacon).
https://www.nbcnews.com/politics/white-house/trump-reportedl...
> A comedian pretending to be Sen. Bob Menendez, D-N.J., says he talked on the phone with President Donald Trump earlier this week in an epic recorded prank call during which the president discussed a range of policy topics.
Bless your soul
That people in government speak like this is utterly absurd. The quote from Donald Trump's follow up tweet on t'social is considerably worse.
This was all due to Antropic not wanting to take on a military contract, right? Or is it suggested its more to do with Mythos, but why would it be, if they never released it.
The substance: traditionally, defense contracts don’t have clauses in them limiting what the military can do with the acquired technology. If Boeing or Lockheed Martin or Northrop Grumann sell a missile system to the Pentagon, they don’t try to impose contractual limits on who the Pentagon can fire the missiles at. Now, for some types of contracts - e.g. contracts to provide personnel - the Pentagon is used to contractual terms limiting uses - but not for hardware or software used in weapons systems / military planning / etc.
Along comes Anthropic, who argue AI is a fundamentally different technology, to which the old rules shouldn’t apply - they want contractual terms prohibiting certain uses (autonomous weapon systems without human in loop; domestic mass surveillance). The Biden admin buys the argument and agrees to those novel contractual terms. The Trump admin takes over and objects to them, demands they be renegotiated. I think it was primarily a matter of principle and power-“software vendors don’t get to tell us what we can and can’t do”-rather than some immediate plan to do things the contract prohibits.
OpenAI negotiated a contract which replicated those terms-but with the proviso that the terms only apply insofar as they reiterate existing legal limits. Anthropic was objecting to that as a meaningless fudge-“we promise not to do X if X is illegal” is very weak, especially when contracting with the government-Congress could change the law tomorrow, or the government’s lawyers could change their interpretation of it, or an appellate court decision could impose a new understanding of it.
It's not like any legally questionable kidnappings or bombing campaigns were being planned at the time, right?
Anthropic’s terms weren’t “don’t do anything illegal” they were “here are two highly specific things which you aren’t allowed to, whether they are legal or not”
They’re arguing it’s a service. I think Aramark could refuse to contract to provide employees to the U.S. military for a campaign on Chicago.
But contracts for personnel generally don’t contain restrictions on use beyond that. If the clerical assistant for DC is asked to provide clerical help to a military planning team who are planning an assault on Chicago, they (and their employer) don’t have legal grounds to refuse. If you are contracted to provide clerical assistance to military planners, you can’t legally say “Baghdad is fine, but Chicago is a no”. Saying that is a breach of contract-unless the courts rule that planning the assault was itself illegal, and I doubt current SCOTUS majority would
Which is what Anthropic was seeking to do. To be clear, I’m unsympathetic to Anthropic’s automated kill-decision objection. But I’m very sympathetic to their no-domestic-use requirement. And that aligns cleanly with precedent for civilian personnel having limits on “the nature and location” of their work.
> Saying that is a breach of contract-unless
Unless you’re negotiating the contract.
One inherent issue with this set of choices is that you're OK with getting killed upon a decision made by a foreign AI agent.
Sadly, such a loophole already exists[0] and means that your own administration wouldn't be spying/executing you with AI, but oops - some ally (not even an adversary!) might on their behalf.
[0]: https://en.wikipedia.org/wiki/Five_Eyes#Domestic_espionage_s...
And then it becomes legal. It’s not an empty argument, it simply means “someone higher than you took an initiative”.
Honestly, Anthropic's stance feels like an oligarch stance. We have better morals than the American people, we will decide what weapons systems the military will use or not use.
It's perfectly understandable if they don't want to sell weapons to the government. That is a noble thing. But Anthropic wanted that DoW money and wanted to determine what is moral vs. not
the people in the current trump admin genuinely believe their own lies
No, they already had a contract (since 2024, revisited/renewed by the Trump admin in mid-2025) which included military usage. That contract, though, had some language about what Claude couldn't be used for, ostensibly because Anthropic was nervous about accuracy in lethal contexts. Hegseth and others were unhappy with the restrictions and wanted to just redo the contract to remove them. Anthropic didn't want that, at least with current models. Then everything blew up. Zvi has some great writeups with more than you probably want to know.