I watched this video this morning, and I intended to share it.

This hammers on one of a few issues that I've had with a lot of the dialogue around fascism. I've said before that Hitler wasn't Hitler until near the end. You wouldn't say he's a Hitler until 1933 at the earliest. And it probably wasn't really until like 1939 or 1941 when people would have started really talking about it like that. Despite that, he wrote Mein Kampf in 1925, staged a coup in 1923 and did a lot of other things before 1939 that were problematic. If Trump was literally doing the exact same things, and was intending to do all of the same things, it would somehow feel inappropriate to make a comparison until several years from now.

The other important thing that I think gets missed is that we have this almost supernatural evil view of Hitler, as if he was the literal anti-christ and no one can ever be that evil. It becomes practically a joke if people compare something to Hitler. It gets dismissed as hyperbole, and people stop listening to you. It is probably actually easier to compare Trump to the devil, than it is to compare him to Hitler.  We talk about Hitler as if he was some almost imaginary nebulous monster, that no one should ever be compared to, and it's impossible to ever be that evil.

And I think that's a huge mistake. I've seen a few tumblr/Reddit posts where Hitler is with a little girl, or a dog, with text that boils down to "Hitler was a person, and this should remind you that people are capable of terrible things". And that's the framework that should really be taken. 

Putin doesn't need to kill 6 million people to be compared to Hitler. Shoving people he doesn't like in concentration camps, and imperialistic push into Ukraine. 

The general devaluing of people who feel differently or don't have the same status, should be enough to call out Nazism. But that feels ludicrous, and we let so much tragedy happen. We convince ourselves that nothing is that bad, until it becomes every bit as bad, at which point it is too late to fix the damage.  

Smoke and mirrors. Sounds like he totally caved, which he did.

Yes, those advancements happen in cycles and sometimes they happen outside the bigger companies you listed… by startups that quickly get acquired with no meaningful pushback, that’s why “the market” isn’t usually spooked.

But a Chinese company that we can’t buy, open source since day 0, a model that can be run locally if desired, this is highly unusual and the closest we have to an actual openAI, ironically. Even when the dust settles, this will always be huge. No wonder so many American CEOs had to say something before they went to sleep.

There are the philosophical aspects to what happened that makes this delicious to watch and why many of us enjoying this even on a subconscious level that we can’t fully articulate. Just few days after all these spineless tech CEOs lined up behind Trump asking for more deregulations, more freedom, less government oversight, and in return, they’ll somehow keep America ahead of everyone else, at least when it comes to AI and silicon valley toys.

But the truth is, China has been competitive at building almost everything. Building impressive infrastructure like ultra highspeed trains, 5G networks, entire high tech cities, manufacturing facilities, automation, making ridiculously cheap useful stuff that I use in every room in my house (and if mandarin was a more popular language they’d probably have a better Hollywood and entertainment content).

It’s as if you can have a government with missions that achieves all of the above while providing affordable healthcare and housing, and now they seem to be doing just fine as well in areas like AI despite all the restrictions? What does a modern-day Western democracy have to say about all of this? 

It turns out that governments that build most of their strategies around pleasing the private sector and big donors are falling behind at everything, (including having the high moral ground and saying no to genocides, and don't get me started on annexing Greenland and childishly renaming gulfs). I'm just rambling now, but this is why the deregulations talks in Europe to “attract growth” these days frustrate me, it’s not better education that we need, it’s not better living the standards, it’s not healthcare that works and actually enables workers to be productive, it’s not affordable housing, it’s not high speed rail connecting cities, no, it’s how we can do what Elon thinks an efficient government should look like. No ambitious missions and long-term plans whatsoever, our solutions are down to pushing up and down the taxes, regulations and consumption levers, pathetic.

off topic but this gave me a much needed chuckle:

edit:

Admittedly, some of the above sounds like hyperbole, to balance things out a little bit, I did like the Biden administration mission to bring manufacturing back and catch up with China (yes, catch up), I also liked Lina Khan and her vision for what competition should look like. And I did want labour to learn from some of that, too bad the same administration didn't take winning a second term seriously.

The Chinese part is what is unusual, yes, although it was always a good possibility Chinese companies would catch up. But open-source isn't unusual. Open-source base-models have been in a "good enough" phase for a while now (including Deepseek's models up until this point.) I've been using them in my job (as an MLE/Data Scientist) for nearly two years, because they allow for more control over data-security in a way connecting to a web-based API doesn't and they are usually (not always, Google has a very cheap API, even cheaper than Deepseek's actually) cheaper.  

For the use-cases that most people are using LLMs for in the business world such as converting unstructured data -> (semi-)structured data, automating the building of knowledge graphs, documentation, transcription, machine translation, text summarization, RPA, etc these smallish base-models have been "good enough" for a while now. 

This is why the cloud-based industry that are trying to sell AI as a service have been moving towards two general goals: 1. building tools that make the base models more useful within an ecosystem (video chat, operators/agents, projects, MCP, etc) -- basically becoming what Google Search is for web-search but for AI, and 2. moving on to reasoning A.I that can help create feedback loops for research.

Most of these large-scale advancements are going into #1. Providing a set of services to hundreds of millions of customers costs a lot of inference and therefore a lot of compute, much more than training the models do, but it is these ecosystems that are going to be monetizable in the cloud (or so these for-profit companies hope.) #2 (outside of OpenAI, as they are the leaders and want to gate-keep) has been a much more collaborative effort between different actors. Anthropic, Google, Nvidia, Meta, Mistral, Deepseek, Microsoft (outside their partnership with OpenAI), Qwen (the other big Chinese open-source group, a subsidiary group of Alibaba), etc all release (to varying degrees) quite a bit of details about what they're doing. #2 belongs to no country and likely won't be monetizable other than the fact it pushes the industry forward. Think of #2 like compiler of old. Compilers used to be primarily proprietary, but eventually open-source won. 

As for yesterday's reaction, I think it is just that a reaction from low-information investors following the herd. Regardless of who is building these models and how efficient they get unless the CUDA-competitor problem is solved, Nvidia is going to be in high demand. More efficient models still follow the neural scaling laws and Jevons Paradox applies here as much as it does to any other basic infrastructure. 

Trump To Tariff Chips Made In Taiwan, Targeting TSMC

💀💀💀💀

White House Pauses All Federal Grants, Sparking Confusion

WTF........ if he's going to tariff anything it should be China.  

There's 2 Scenarios.

1. Typical Trump bluster, he makes threats and big demands, so the country or company come to the table to talk to the baby, they give the baby a minor win but the baby sells it to his dumbass fanbase as a major win when in actual fact, not much has changed at all, his fanbase claps like seals and shout loudly about how much of a genius he is.

2. He's a bigger idiot than we all possibly imagined and is about to utterly destroy America's economy and make everything extremely expensive for everyone, it takes literally years, upon years, to build chip factories, none of which will be felt in his term but what will be felt is him destroying America's economy to the average person.

Lets hope it is #1. I still believe the Canada and Mexico tariffs might be, hence why he didn't issue them on day one but set a date of Feb 1st. I still believe there's a chance he will be like "Guys, I got an amazing deal with Canada and Mexico, the best deal you've ever seen, the issue is resolved" and his dumbass fanbase won't bother questioning or looking too deeply into it and believe him as per usual but in actual fact, nothing much has changed but Trump loves stroking his ego more than anything else so he loves this "create a fake issue, fix the fake issue" scenarios.

If it's #2 then God help America, Lol.

At least egg prices are coming do-

Egg prices are soaring. Don't expect that to change anytime soon | AP News

Oh darn. RFK Jr to the rescue! Lmao.

I'm in the middle on this between complete lunacy/idiocy on one hand, and a complete win for America.  The move just seems odd given Trump's repeated insistence that China is the new big and growing threat towards America.  I think that he is wanting America to start building more chips, but this is a poor way to accomplish that.  He should be giving generous subsidies and so forth, not fighting against a weak nation that may get destroyed by China in the future.  

China is going to be the biggest winner of these next 4 years, Trump may get his temporary PR wins from time to time but in the long-term it will have a horrible impact on America's soft power/global influence, which MAGA is too stupid to understand the importance of. Threatening allies 24/7 will only drive them to move away from you in the long term because you're a shit ally, a shitty trading partner, an abusive bully.

Threatening allies and trading partners across the world will push countries to seek allies elsewhere, both in terms of military support and increasing trade with others, Trump is now the biggest advertiser of BRICS and the dumbass thought Spain was apart of BRICS because of the S in the name. Even those who placate the baby in the short term will likely seek greener pastures in the long term.

European unity and independence from America will long-term be a net positive for us.