They could easily take the weights if they wanted. I don't believe they meaningfully lost access.
HlessClaudesman 22 minutes ago [-]
Who will make them the next set of weights?
If a government can just seize the product of someone else's labour, either they will end up as slave owners or without willing workers.
dofm 16 minutes ago [-]
Serious question: do you think the NSA aren't training their own LLMs? (With or without Anthropic and OpenAI's help)
It's a perfect technology for their uses, they get a big chunk of a $100 billion black budget, and they've had access to the research for at least as long as we have.
ben_w 8 minutes ago [-]
> Serious question: do you think the NSA aren't training their own LLMs?
Given the evergreen discussion of "are these companies making a profit"*, I think any LLMs that the NSA (or any other government agency worldwide) may be making are quite far from the leading edge.
* Person A: "they are making a loss!" Person B: "Only if you count training, they make a profit on inference, look at what it costs to run comparable open models on generic cloud servers" A: "Sure, but if they don't train new models they'll be left behind, so they're still making a loss"
That and the way compute is now measured in GW, I think even random low budget vloggers just getting started would be able to spot if the NSA was doing anything significant just from the extra heat emissions or power plants getting built.
ACCount37 3 minutes ago [-]
Model training does NOT dominate the model costs.
The rate of inference compute to training compute is ~10:1, for popular frontier models. Models are routinely overtrained past the Chinchilla optimum now because it makes an immense amount of economic sense to do so.
Worse the more niche and unused your models get, but when this "making a loss" fuckery pops up, it's usually about the big guys like Anthropic, OpenAI, GDM and maybe xAI and Meta. Of which only the latter can be accused of not selling enough inference to offset the training runs.
The real money sinks are: R&D and infrastructure buildouts.
HlessClaudesman 13 minutes ago [-]
I don't think there is much overlap between people capable of building cutting edge LLM's and the people who want to build a cutting edge LLM for the government.
dofm 4 minutes ago [-]
The NSA managed to deliberately insert a backdoor into elliptic-curve cryptography right under the noses of everyone capable of making elliptic-curve cryptography.
I wouldn't count them out.
dgellow 2 minutes ago [-]
You cannot really hide the amount of compute required to train an LLM. Do we have actual clues that NASA is training their own frontier model?
polytely 8 minutes ago [-]
They probably also have an insane dataset
segmondy 51 seconds ago [-]
Serious question, do you realize that the NSA are mere mortals? Do you realize how much it takes to train a model? Does the NSA make their own chips or planes? The NSA buys a lot of technology because they can't make their own.
stronglikedan 6 minutes ago [-]
> do you think the NSA aren't training their own LLMs?
They probably already have access to Sentinel, so they wouldn't need to train their own.
infinite_spin 20 minutes ago [-]
the success of mythos isn't from model weights, it's from the harness and toolset it has access to
krzyk 13 minutes ago [-]
Is it really?
Harness is important for model performance, but weights are surely mode important, without that you would have haiku doing the work.
dofm 18 minutes ago [-]
I agree but that's even easier to exfiltrate, surely.
47 minutes ago [-]
Onavo 37 minutes ago [-]
If they use the defence production act, would Dario be even able to resign in protest?
AustinDev 35 minutes ago [-]
If they wanted to officially take the weights the DPA would work and Dario could do nothing. If they wanted to do it in clandestine manner no one could stop them and no one would know. It's very likely they already have all the weights from all the frontier models. I mean all the frontier models are capable of being served from AWS Bedrock so the weights aren't exactly locked in some air-gapped vault.
It would be easy to make a national security justification to take the weights in a clandestine manner especially because Anthropic supposedly got caught giving China access to the model through a cutout.
rurban 6 minutes ago [-]
John Cook resigned, so Dario might resign also. But he would make it public, so they won't do it
5 minutes ago [-]
wetpaws 41 minutes ago [-]
[dead]
dofm 21 minutes ago [-]
If Mythos is still running internally, the NSA still have some access to it. It's just crazy to believe there aren't CIA and/or NSA plants (tacitly acknowledged or otherwise) inside Anthropic and OpenAI.
But Mythos is still only an advanced LLM so I am not sure what all this breathy fuss is about; it sounds like the PR war more than anything.
If the NSA aren't themselves training technologies that are at least as powerful, that would modestly surprise me.
Not that you need an LLM to monitor the risks to the USA. You just need Tulsi Gabbard's emails.
SV_BubbleTime 11 minutes ago [-]
I think it’s beyond a mastery of PR. They literally called it Mythos and built a literal myth around it. I mean… maybe people just want the soap opera.
charcircuit 3 minutes ago [-]
Everyone lost access. What even is mainstream news these days.
ck2 46 minutes ago [-]
they are doing DOGE-cuts to all of intelligence now anyway
dozens upon dozens fired for no reason
so US "intelligence" is going to go even further backwards
The NSA is managed by the NSA director, an independently appointed and confirmed office separate from the Office of the Director of National Intelligence. The DNI does have the authority to restrict funds to the NSA, and sets certain priorities, policy, etc. but the NSA director is not beholden to the DNI and makes their hiring and firing decisions independently. They’re also, currently and historically though not required by statute, a flag officer in the US military and dual-hatted as the commander of CYBERCOM. All this is to say, chaos in the office of the DNI does not necessarily impact the NSA.
kansface 36 minutes ago [-]
Here is one sector of the US government I'm happy to see burned down. If the alternative is the status quo, I'm OK with any roll of the dice.
tempodox 24 minutes ago [-]
It can always get worse.
sailfast 33 minutes ago [-]
Anecdata suggests NSA just got on board and kept going tbh. Not sure they’ve felt the same impacts / churn as other agencies, and not sure they’ve ever really been that beholden to the DNI.
parineum 40 minutes ago [-]
> dozens upon dozens fired for no reason
When you say without reason do you mean without cause?
islandfox100 23 minutes ago [-]
Seems to me OP's implication is that they were fired because someone wanted to hit a quota of (employees cut/payroll expenses reduced), or other similarly ''reasonless'' justifications.
medlazik 41 minutes ago [-]
AI marketing bullshit stunts are unlike anything I've seen in 30 years. It started with MS Copilot so called capabilities for work, which were completely made up use cases that didn't work at all (3 years later still). We've had OpenAI "AGI is coming" and "AI will take your job", now we have Mythos being so "dangerous" for cybersecurity, which of course makes the average Joe interpret it as Anthropic being "the better overall company, the NSA uses it!!". I mean gov foes with Anthropic are probably true, but the marketing is to blame not Mythos capabilities. This is all so fucking pathetic
thewebguyd 23 minutes ago [-]
> and "AI will take your job"
Don't forget, its no longer cool to say that now that the public has pushed back. The fact they all changed their tone away from taking jobs tells you that it was all just entirely marketing.
scottyah 6 minutes ago [-]
Seems to me that they were mostly right, and the message was received by the right people. No need to ensure it gets distributed to the wrong people.
tempodox 7 minutes ago [-]
But the propaganda deluge was a smash hit so far, HN is drowning in “AI” BS, and astroturfers and spin doctors haven’t seen that much business since the cold war. They made more profit than shovel salesmen in the gold rush.
colechristensen 30 minutes ago [-]
I was able to identify, diagnose, fix, and upstream a minor bug in and erlang/OTP ssh key implementation with Opus in maybe 20 minutes (+2 weeks or so for upstream). It is not impossible that I could have done this before, but it would have taken days or weeks. The actual fix was about 2 lines of code, hardly AI slop, but getting there would have been quite the slog, and I never would have done it.
There is a lot of the reason for AI skepticism out there, but people tend to do massive overcorrections and underestimate the force multiplier it can be, particularly for people with some idea of what they're doing and a good grasp of how to take advantage of the tool.
medlazik 18 minutes ago [-]
I said absolutely nothing about LLMs, which is a fantastic tool I'm using every day. I'm talking about marketing.
gallerdude 3 minutes ago [-]
So let’s say you’re in Anthropic’s shoes. You see that LLM’s are getting better and better, and it’s very possible that they will have some impact on jobs in the next few years, and a very meaningful impact on cybersecurity.
Is it more ethical to stay silent about these concerns, as you might have a bit of self interest? Or even if it looks a bit self interested, is it better to warn people ahead of time? I think the latter is obviously the better position.
bflesch 35 minutes ago [-]
We should seriously reframe this whole AI thing to "SI = simulated intelligence".
It's google in a box. Great achievement, makes knowledge work faster, but please stop bothering everyone else.
The Uber and Groupon people became billionaires, so the "Simulated Intelligence" folks will also achieve it. No need to worry and drown everyone in these bs stories only non-tech people believe.
13 minutes ago [-]
AnimalMuppet 21 minutes ago [-]
Heh. In the Schlock Mercenary universe, "SI" means "synthetic intelligence", which is a level below real AI (which means what we would call AGI). And, as it says (in https://www.schlockmercenary.com/2003-07-21), SI translates to "kinda stupid".
ToucanLoucan 38 minutes ago [-]
All for a product that has yet to make a single honest dollar in profit for anyone who isn't nvidia.
When this goes we might well see a recession. Not that anyone responsible will be worse off, of course.
scottyah 5 minutes ago [-]
Why on earth would you expect any of them to take profit so early in the game?
tempodox 20 minutes ago [-]
The perpetrators all have their golden parachutes. The taxpayers will foot the bill.
micromacrofoot 43 minutes ago [-]
Absolute clown show, wonderful for Anthropic to keep themselves in the news cycle yet again.
tmaly 1 minutes ago [-]
there is a well known politician that uses this same trick
Madmallard 1 hours ago [-]
Doesn't make any sense. They could just force them to provide Mythos to the federal government.
JumpCrisscross 1 hours ago [-]
> They could just force them to provide Mythos to the federal government
The DPA only gives that power to the President [1].
Which is a fundamental mistake to make with the U.S. government, even if we’re talking only about the executive branch, even if we’re only talking about DoD, even if we’re only talking about the IC.
flybarrel 20 minutes ago [-]
doubt Trump would accept that pronoun
graemep 60 minutes ago [-]
The current position seems to be no-one has access, not even Anthropic employees. What powers does the US government have to force them to provide access? If they have that power why did they not use it to force them to provide their products for military use?
ceejayoz 50 minutes ago [-]
> What powers does the US government have to force them to provide access?
This would not be a particularly big stretch here, either.
graemep 44 minutes ago [-]
There is even a precedent for its use with regard to AI (only disclosing information, but still). Biden used it, why does Trump not do so?
aleqs 1 hours ago [-]
Yeah... NSA literally has MITM proxies/interception of any traffic they want inside every major US tech company (based on my reading/following of Snowden leaks and others). Anthropic wouldn't be able to exist without implicit NSA approval. This article reads more like a marketing piece for Anthropic/Mythos... and ends by talking about how much NSA wants Anthropic models.
How are they going to MITM communications with certs that never left my machine?
Are you suggesting they broke TLS or that they've somehow acquired every private cert generated?
aleqs 27 minutes ago [-]
You just intercept the traffic after its decrypted on the server side, or are you suggesting you somehow send encrypted traffic that never gets decrypted?
vintermann 51 minutes ago [-]
It's back to the question of how much you should give the benefit of doubt to powerful people who openly lie.
strictnein 41 minutes ago [-]
It's just not technically feasible, so there's nothing to lie about. They're not MITMing petabytes/sec across dozens (hundreds?) of companies and they haven't broken TLS1.3.
If I have a box at Digital Ocean and I'm communicating with it with TLS1.3 using a Let's Encrypt cert that I generated, where, exactly, does this magical MITM box come into play?
drdexebtjl 16 minutes ago [-]
That "box" is a virtual machine, no?
Do you know what hypervisor is managing it? :)
aleqs 24 minutes ago [-]
Of course it's feasible, you just intercept the traffic post-decryption on the cloud/server side. You don't control how/where your traffic to 3p cloud services is decrypted.
chews 46 minutes ago [-]
bruh, it's not speculation. The secret NSA surveillance room in San Francisco is officially known as Room 641A. It is located inside an AT&T switching center at 611 Folsom Street (near 2nd Street) in the SoMa neighborhood. Who else occupied the building, Twitter was also located at 611 Folsom Street in San Francisco.The company leased a floor in the building (sharing space with retro platform Bebo) in 2009. This was their third office space, serving as their headquarters before they outgrew it and eventually relocated to Market Square at 1355 Market Street in 2012. The arab spring twitter uprising was fully a CIA/NSA operation.
XMPPwocky 31 minutes ago [-]
> This was their third office space, serving as their headquarters before they outgrew it and eventually relocated to Market Square at 1355 Market Street in 2012. The arab spring twitter uprising was fully a CIA/NSA operation.
To be clear, the claim you're making is that because Twitter has their third corporate office in the same building as an AT&T switching center, and US intelligence used a room in AT&T's switching center for surveillance, then Twitter must have been controlled by US intelligence? And thus the Arab Spring uprising, where Twitter was used, was "fully a CIA/NSA operation"?
chews 7 minutes ago [-]
Yes, twitter was used by US 3 letter agencies to assist in the arab spring. To be able to do it in a surreptitious way they were asked to move to that building and get access to all private DMs, and for doing so they got a fat tax break to move to "blighted" market street current location. All of those things fit the timeline and snowden capability disclosures.
Yes, you have collected a lot of random bits of information from over a decade ago. I'm sure everything you say is still relevant today, especially the conspiracy nonsense.
Some of us actually work in security, while others think the NSA and CIA are some magically powerful orgs.
Explain how, even with the mystical Room 641A, the NSA can't break a TLS1.3 protected communication channel without either party knowing about it. Assume you have generated a cert with Let's Encrypt. How, exactly, does that work?
aleqs 22 minutes ago [-]
Explain to me how you are going to encrypt your LLM API calls with your let's encrypt cert.
There are also multiple ways/places traffic you send to typical cloud/tech company is decrypted and can be intercepted. (Surprised I have to point this out to someone who 'actually works in security ' lol)
Not to mention US tech companies fully cooperate with the NSA in many cases and are aware of this going on.
sailfast 30 minutes ago [-]
Thank you.
wan23 1 hours ago [-]
You misunderstand - the government issued a directive to Anthropic that effectively forced them to pull access from everyone, even their own employees.
hk__2 1 hours ago [-]
The directive was to remove access to non-Americans, not to pull access from everyone. It’s because Anthropic cannot verify the identity of its users that it pulled access from everyone, not because the government explicitely requested that.
greatpatton 47 minutes ago [-]
If their operation team is not US based that's going to be difficult to operate. They would have to reorganize the whole company as I'm pretty sure that they are not employing only US citizen.
msm_ 34 minutes ago [-]
>I'm pretty sure that they are not employing only US citizen
Understatement. They have 14 offices, only 4 of them are in the US (6 are in EMEA, 4 in APAC).
JumpCrisscross 46 minutes ago [-]
> directive was to remove access to non-Americans
Did Hegseth pull his supply-chain risk BS?
sailfast 29 minutes ago [-]
No - this was a separate power unrelated to the supply chain risk which is still in effect.
bluGill 59 minutes ago [-]
Probably not. The US constitution limits what government can force on the people. If the NSA tries to force something that will spend years in court (if anyone wants to fight)
folkrav 50 minutes ago [-]
The constitution limits a lot of things that this administration has done regardless.
stackghost 20 minutes ago [-]
> The US constitution limits what government can force on the people.
The US constitution also prohibits:
- refusing to spend money that congress has appropriated
- dismantling congressionally-created federal agencies without congressional authorization
- directing federal agencies to selectively apply the law according to the preference of the executive
- giving control of federal agencies to individuals who have not been appointed by the legislative branch
- terminating, detaining, or deporting people without due process
- retaliation against private citizens or corporations for speech protected under the first amendment
- discriminating on protected grounds under the equal protections clause
... and yet the administration has done all these things with impunity while effete judges wring their hands and write sternly-worded letters. The US constitution demonstrably no longer has any force or effect.
gsibble 1 hours ago [-]
Misdirection
swader999 49 minutes ago [-]
Kind of crazy actually. Other models are catching up fast, they all can find the vulnerabilities in our (and by our I mean everyone's) underlying infra very fast. It takes a very long time to fix, review, and finally deploy these fixes. There really isn't much time left.
If a government can just seize the product of someone else's labour, either they will end up as slave owners or without willing workers.
It's a perfect technology for their uses, they get a big chunk of a $100 billion black budget, and they've had access to the research for at least as long as we have.
Given the evergreen discussion of "are these companies making a profit"*, I think any LLMs that the NSA (or any other government agency worldwide) may be making are quite far from the leading edge.
* Person A: "they are making a loss!" Person B: "Only if you count training, they make a profit on inference, look at what it costs to run comparable open models on generic cloud servers" A: "Sure, but if they don't train new models they'll be left behind, so they're still making a loss"
That and the way compute is now measured in GW, I think even random low budget vloggers just getting started would be able to spot if the NSA was doing anything significant just from the extra heat emissions or power plants getting built.
The rate of inference compute to training compute is ~10:1, for popular frontier models. Models are routinely overtrained past the Chinchilla optimum now because it makes an immense amount of economic sense to do so.
Worse the more niche and unused your models get, but when this "making a loss" fuckery pops up, it's usually about the big guys like Anthropic, OpenAI, GDM and maybe xAI and Meta. Of which only the latter can be accused of not selling enough inference to offset the training runs.
The real money sinks are: R&D and infrastructure buildouts.
I wouldn't count them out.
They probably already have access to Sentinel, so they wouldn't need to train their own.
Harness is important for model performance, but weights are surely mode important, without that you would have haiku doing the work.
It would be easy to make a national security justification to take the weights in a clandestine manner especially because Anthropic supposedly got caught giving China access to the model through a cutout.
But Mythos is still only an advanced LLM so I am not sure what all this breathy fuss is about; it sounds like the PR war more than anything.
If the NSA aren't themselves training technologies that are at least as powerful, that would modestly surprise me.
Not that you need an LLM to monitor the risks to the USA. You just need Tulsi Gabbard's emails.
dozens upon dozens fired for no reason
so US "intelligence" is going to go even further backwards
* https://www.yahoo.com/news/politics/articles/trump-acting-ch...
November is going to be insanity
When you say without reason do you mean without cause?
Don't forget, its no longer cool to say that now that the public has pushed back. The fact they all changed their tone away from taking jobs tells you that it was all just entirely marketing.
There is a lot of the reason for AI skepticism out there, but people tend to do massive overcorrections and underestimate the force multiplier it can be, particularly for people with some idea of what they're doing and a good grasp of how to take advantage of the tool.
Is it more ethical to stay silent about these concerns, as you might have a bit of self interest? Or even if it looks a bit self interested, is it better to warn people ahead of time? I think the latter is obviously the better position.
It's google in a box. Great achievement, makes knowledge work faster, but please stop bothering everyone else.
The Uber and Groupon people became billionaires, so the "Simulated Intelligence" folks will also achieve it. No need to worry and drown everyone in these bs stories only non-tech people believe.
When this goes we might well see a recession. Not that anyone responsible will be worse off, of course.
The DPA only gives that power to the President [1].
[1] https://en.wikipedia.org/wiki/Defense_Production_Act_of_1950
https://en.wikipedia.org/wiki/Defense_Production_Act_of_1950
This would not be a particularly big stretch here, either.
Propaganda.
https://www.wired.com/2013/10/nsa-hacked-yahoo-google-cables...
https://www.eff.org/nsa-spying
IPO incoming.
No, they don't.
https://en.wikipedia.org/wiki/Room_641A
Yeah, they did (and probably do).
Are you suggesting they broke TLS or that they've somehow acquired every private cert generated?
If I have a box at Digital Ocean and I'm communicating with it with TLS1.3 using a Let's Encrypt cert that I generated, where, exactly, does this magical MITM box come into play?
Do you know what hypervisor is managing it? :)
To be clear, the claim you're making is that because Twitter has their third corporate office in the same building as an AT&T switching center, and US intelligence used a room in AT&T's switching center for surveillance, then Twitter must have been controlled by US intelligence? And thus the Arab Spring uprising, where Twitter was used, was "fully a CIA/NSA operation"?
The CIA venture arm InQTel invested in Dataminr a company that twitter was also a major shareholder. https://theintercept.com/2016/04/14/in-undisclosed-cia-inves...
Some of us actually work in security, while others think the NSA and CIA are some magically powerful orgs.
Explain how, even with the mystical Room 641A, the NSA can't break a TLS1.3 protected communication channel without either party knowing about it. Assume you have generated a cert with Let's Encrypt. How, exactly, does that work?
There are also multiple ways/places traffic you send to typical cloud/tech company is decrypted and can be intercepted. (Surprised I have to point this out to someone who 'actually works in security ' lol)
Not to mention US tech companies fully cooperate with the NSA in many cases and are aware of this going on.
Understatement. They have 14 offices, only 4 of them are in the US (6 are in EMEA, 4 in APAC).
Did Hegseth pull his supply-chain risk BS?
The US constitution also prohibits:
- refusing to spend money that congress has appropriated
- dismantling congressionally-created federal agencies without congressional authorization
- directing federal agencies to selectively apply the law according to the preference of the executive
- giving control of federal agencies to individuals who have not been appointed by the legislative branch
- terminating, detaining, or deporting people without due process
- retaliation against private citizens or corporations for speech protected under the first amendment
- discriminating on protected grounds under the equal protections clause
... and yet the administration has done all these things with impunity while effete judges wring their hands and write sternly-worded letters. The US constitution demonstrably no longer has any force or effect.