However, thinking back to the spat with the DoD and more generally how the administration is much more supportive - and supported by - OpenAI and XAI and it’s easy to imagine this is just another escalation in the fight between a “liberal leaning” company and its competitors and the administration.
There might have been something said by someone at Amazon or something but I’d guess Occam’s razor the administration just leapt at the chance after their supplier sanctions fell flat?
Then the companies realize fighting the US government is a lot of effort, expensive and creates a lot of drama, and it's easier to reach a mutual understanding.
Have a legal department trained to be the buffer between you and the government any contact any questions goes through them and since you’re paying them a ton of money, they are the only people the politicians should get to know, oh, and again and again do not volunteer anything.
Considering how miniscule their ideological demands were, I think this is more of a classic "any amount of push back, no matter how miniscule will be punished to achieve full power" then anything to do with "Ideological Resistance".
It may be due to principles and the principled stand Anthropic has taken has caused trouble, but they have also not delivered a bribery as far as we can see.
When there is a protection money racket in play, it’s hard for me to take other factors as seriously. This is gangster economics not a philosopher’s circle.
Here’s another take on this with more details: https://news.ycombinator.com/item?id=48593212
By which I mean, they are both interested in being filthy rich, making hay while the sun shines, and stopping centrist forces from adding wealth taxes or anti-monopoly regulations.
Biden and Harris were getting too uppity about such things for their tastes.
Have you seen who Sergey Brin is budding around with these days, and what he -- a formerly very "left liberal" person involved in very vocal anti-Trump protestations inside and outside of Google -- is saying these days? Third richest dude in the world is terrified of wealth taxes being put in place by his former friends, and dating a MAGA "gut-health" influencer.
A gentleman's agreement was brokered between SV and the GOP involving some mutual backscratching.
Which Dario and crew aren't a part of. Yet.
"Cash rules everything around me, CREAM, get the money, dollar dollar bill, yall"
Are you joking? When was this? Remember, Microsoft was the very first partner to deliver working warrantless surveillance capabilities (against their customers) to the NSA as part of the unconstitutional PRISM program, and this was back in 2007, nearly two decades ago.
For me Occam's Razor are these facts:
1. The Chinese are distilling American models
2. Anthropic last models have national security implications (large volume of exploit creation)
3. China is the US major adversary today
That's why restricting access to the models sound like a reasonable to do
No dog in this fight but the Occam’s razor here is the story, model could be tricked to do things it was hobbled from doing
Mythos is clearly dual use (and automatically subject to export controls), even if Anthropic didn’t understand what that means.
Yes this administration can be capricious and hyperpartisan. This isn’t one of those cases.
It may have been a contributing factor, but the crux of the shutdown was the industry reporting of Fable jailbreaks (reportedly spearheaded by Amazon CEO Andy Jassy). The more interesting and honest angle is that the industry which has taken the seriousness of Glasswing at face value felt blindsided by Fable release and totally exposed by the residual risk, when they know they still have a months-long bugfixing backlog exposed by Glasswing and are desperate to buy more time.
This misleading looks deliberate on Wired’s part, to appear as though they’re getting a scoop when they’re really just being dishonest. Shameful.
Two problems with this theory.
1. Amazon complaining to the White House wouldn't have been the opening salvo. Amazon and Anthropic would find it much easier to talk to each other than go through the White House. We'd need evidence that Amazon (and probably others) already asked Anthropic to not release a Mythos-class model but Anthropic released it anyway. Are they on record saying this?
2. The jailbreak Amazon found needs to be real. Maybe the White House staffers are not AI experts and they don't really understand what a jailbreak is... but it's much harder to make that claim about Andy Jassy. For the jailbreak to be the real reason for the export control order, the jailbreak would need to be significant and cause material harm to Amazon. Then Jassy might pass it along to the White House assuming he already was refused by Dario.
But there is no evidence the jailbreak was real. There is one story that it amounted to a request, "fix this code." In any case, Anthropic is on record saying the so-called jailbreak didn't enable any vulnerability work that couldn't already be done by other models.
The thinking traces disappeared because Anthropic changed them to be hidden by default. The rationale for hiding it was that most people don't look at the thinking traces https://news.ycombinator.com/item?id=47664442 . You can reenable thinking traces in Claude code settings with the flag showThinkingSummaries: true.
Why do you think China will?
I am quite certain the gap will only grow
> the White House asked Anthropic to revoke SK Telecom’s access to Mythos, according to a person close to the AI lab. The company immediately complied,
Lesson learned - don't invest in US companies
I'm sure they're not happy about losing access to the model, but the amount of money they're going to make from their investment will more than make up for it.
Also if you have an agreement with a company for them to provide you with a service, and investing in them is part of that deal, reneging on the service part still isn't okay.
Not if the USG locks models down to US citizens. The market will be too small and the model companies have already pumped in far too much money to limit their market to US citizens. Given that most big companies have a global presence they're going to need models that all of their employees can use. They're not going to deploy different products to different employees.
If you invested in the company, you are a part owner.
If you are part owner, you deserve to have access to any company internals.
Like who ? I find your statement to be doubtfyl
there are many Chinese manual laborers in Korea, Chinese capital has also been heavily integrated into top-tier Korean tech firms and conglomerates. For example, Tencent is one of the largest shareholders in Korea's biggest gaming and tech companies, holding massive stakes in Kakao, Netmarble, and Krafton, while recently attempting to acquire major stakes in Nexon and Kakao Mobility. To say they are rarely 'significant business partners' is factually incorrect. This is just the tip of the iceberg the portfolio of Chinese capital is diverse and expands across media including JTBC which has been pushing very pro-beijing and anti-american view in particular framing the current bipartisan wariness of China as "far right MAGA" exactly as you typed. lee Jae-myung administration which generally leans more toward Beijing has to carefully navigate. Non-partisan groups like Pew Research consistently report that roughly 80% of South Koreans hold an unfavorable view of China making it one of the highest in the world. Notably, Pew also found that South Korea is unique in that its younger generation holds more negative views of China than its older generation the exact opposite demographic makeup of the US MAGA movement.
Dismissing these well-documented geopolitical and economic realities as a fringe talking point simply doesn't reflect the situation on the ground in South Korea and this is why I ask,
are you really Korean and how come you don't know these basic facts? a quick search in Korean would reveal numerous articles that back up China's capital deployment in South Korea especially in sensitive areas.
you can't really fault Koreans for their wariness, many Chinese pretend to be Korean online/offline without good intentions. I can tell you that this is only going to end in violence and more discrimination against Chinese in Korea and people who don't even support CCP are going to be impacted.
https://www.reuters.com/world/asia-pacific/china-restricts-o...
It's going to be interesting to see how the AI arms races unfolds over time.
A student wanted to create a project that bypassed the GF of China. He basically told him he was an idiot, it was impossible, and discouraged the student from even trying. This was a hacker (the security leader) at one point in his life, that claimed to support the freedom of speech and expression.
The actual reason he said this to the student is because he supports the government in China. Many other comments I heard after this proved to me that this is true. My guess is that his startup even received funding from the Chinese government.
It's really disappointing that tech leaders can publicly support murderous regimes like the Chinese government for decades with no blow back. They are far worse than Putin/the Russian government and have 100X the money and technology.
If he supported Russia, he would probably get cancelled and never be able to be involved in the tech industry again.
Anthropic may discover a lack of availability for HBM memory in the near future.
