Which brings the question, if LLMs are an asset of such strategic value, why did China allow the DeepSeek to be released?
I see two possibilities here, either that the CCP is not that all-reaching as we think, or that the value of the technology isn't critical, and that the release was further cleared with the CCP and maybe even timed to come right after Trump's announcement of American AI supremacy.
I really doubt there was any intention behind it at all. I bet deepseek themselves are surprised at the impact this is having, and probably regret releasing so much information into the open.
It is hard to estimate how much it is "didn't care", "didn't know" or "did it" I think. Rather pointless unless there are public party discussion about it to read.
It will be assumed by the American policy establishment that this represents what the CCP doesn't consider important, meaning that they have even better stuff in store. It will also be assumed that this was timed to take a dump on Trump's announcement, like you said.
And it did a great job. Nvidia stock's sunk, and investors are going to be asking if it's really that smart to give American AI companies their money when the Chinese can do something similar for significantly less money.
I mean, it's a strategic asset in the sense that it's already devalued a lot of the American tech companies because they're so heavily invested in AI. Just look at NVDA today.
This seems more like a move designed to frighten China -- or force them to spend money making LLMs -- then an actual threat. The clues are that Trump ceremonially blessed the deal but did not promise money (SoftBank et al will, supposedly), and then Musk said that's all fake because SoftBank doesn't have the money, and Altman countered that Musk should not be butthurt and should put America first. Who does that? I'm thinking, no one who has something real on his hands.
Jesse Schell's book is a great read beyond game design.
Thanks for the other links.
To leave something in return, here's something I read the other day and kept thinking about it (I'm designing on a PvP motion based game)
"In competitive games, there is little more valuable than knowing the mind of the opponent, which the Japanese call “yomi.”
As a side note, I would even argue that the “strategic depth” of a game should be defined almost entirely on its ability to support and reward yomi."
The Yomi Layer concept is a reminder that moves need to have counters. If you know what the opponent will do, you should generally have some way of dealing with that.
I have tons of friends that took his classes at CMU, as much as everything he says sounds good I don't know a single person that has ever enjoyed a game he made. Because of that, I have to assume what he says is either fluff or wrong even if i can't perceive why exactly
That's interesting, it hasn't occurred to me to check his games. That said, I remember reading that Machiaveli was once given a territory to govern and he was terrible at it, despite The Prince. It may be a thing about teachers vs doers.
THAT said, there is a lot of intersting things one can learn from John Carmack, so there's an exception to every rule.
Obviously you’d be a better judge than I am given your inside knowledge, but your comment reminds me of the many virtuoso musicians who don’t really make music people like. There’s a gap between technical ability and taste, artfulness.
It's not that he's making games too avant garde or something that might be going over my head. His company just makes corporate slop, not a single enjoyable game amongst them
Usually but not always. There have been some great corporate projects, but it does feel like they are less common now than say 10+ years ago, and typically neat twists on existing successes vs. unique ideas
Another adage is "code should be written for people to read, and only incidentally for machines to execute". This goes directly against code being written by machines.
I still use ChatGPT for small self-contained functions (e.g. intersection of line and triangle) but mark the inside of the function clearly as chat gpt made and what the prompt was.
> You don’t need unit tests if you have integration tests.
Which is why, as per Jim Coplien, most unit testing is waste.
But converting one type of unit tests into another is a perfect showcase for AI-generated code. They could have even kept just the prompts in the source and regenerate the tests on every run, were it not for inaccuracy, temperature, and the high cost of running.
On the internet, no one hears you being subtle. (Torvalds)
I'll add my own view: when you watch a movie, read a book, listen to a song, play a game... you CONNECT with the mind of the person who made it. When there is no mind, or the source is a dead, statistical amalgamation of countless fragments of other minds, there is nothing you'll want to connect to, nothing you'll want to squander precious hours of your life on.
And while you may be curious to see, once maybe, a movie such an imaginary AGI-LLM has created from your prompt, no one else will have the slightest interest in seeing it. And vice versa. Which means there would be absolutely NO MONEY in that market. There would be no market.
I find it draining to have to be on the lookout constantly for hallucinations, or omissions a person wouldn't make. I imagine as long as I'm walking well known paths -- well known to many but not me -- I'm safe, but the moment I need nuances I can expect that one of those nuances is completely and convincingly made up, except I don't know which one.
I see two possibilities here, either that the CCP is not that all-reaching as we think, or that the value of the technology isn't critical, and that the release was further cleared with the CCP and maybe even timed to come right after Trump's announcement of American AI supremacy.