oytis

28 ms·

661.
by oytis 1y ago
I'd bet AI is used there as well
662.
by oytis 1y ago
Like, to write grant reports? That's already happening on scale.
663.
by oytis 1y ago
Same in Germany. Just accessing data you are not supposed to access is a crime, regardless of whether you used the data in any illegal ways.
664.
by oytis 1y ago
Those not put off will not buy an EV anyway
665.
by oytis 1y ago
Same value and same work as today I guess?
666.
by oytis 1y ago
I am not thinking of a scenario where all labour is eliminated, but just knowledge work. In this case one can divest from most tech companies (except those running AI infrastructure), and reinvest in something else. Tech CEOs will not be ne
667.
by oytis 1y ago
I have this impression as well. AI, interest rates are secondary. Tech industry has just run out of ideas that would promise profits - and that didn't start in 2023. Same story as with "overhiring". If you have a bigger crew,
668.
by oytis 1y ago
Yeah, if workers are not needed, CEOs are not needed either. Luckily most CEOs are rich enough to also be investors.
669.
by oytis 1y ago
They are counting people per company per unit of time
670.
by oytis 1y ago
As much as I'd like it, it would be inconsequential for YCombinator's portal to ban AI-generated posts.
671.
by oytis 1y ago
OK, so we are having two classes of problems here - ones worth solving unreliably, and ones that are better solved without LLMs. Doesn't sound like a next level of abstraction to me
672.
by oytis 1y ago
I don't get his argument, and if it wasn't Martin Fowler I would just dismiss it. He admits himself that it's not an abstraction over previous activity as it was with HLLs, but rather a new activity altogether - that is promp
673.
by oytis 1y ago
It was a joke back then, not a real business model
674.
by oytis 1y ago
Microsoft? They do advertising among other things, but I don't think it makes the largest chunk of their revenue
675.
by oytis 1y ago
That's an improvement, they are still "chaotic" though in that small changes in input can change the output unpredictably strong
676.
by oytis 1y ago
But that's essentially the bright future drawn by American AI companies. As on copyright vs no copyright, I agree with the view of Bunny Huang. Chinese laxity with copyright doesn't destroy the advantage of the first mover - it ju
677.
by oytis 1y ago
What about those who want specifically AI oligopolies to fail?
678.
by oytis 1y ago
From the snippet I can read looks like open source is a authoritarian thing
679.
by oytis 1y ago
Can someone working in the field comment on how relevant the content still is? A lot of ML including CV seems (from the outside at least) to be completely disrupted by the developments of the last two years.
680.
by oytis 1y ago
I am not reading it like this - in fact ChatGPT was the first thing out there that would be able to assist them in writing, and less than a third of this book was written after release of ChatGPT. To me it just looks like marking important
681.
by oytis 1y ago
Other people are different, that's the thing, while AI is generally predictable quality, and it's not going to go down. Autonomous driving is just one example, I really think it's a general pattern
682.
by oytis 1y ago
People are eager to pay money to not deal with other people. Which makes me pessimistic about the future of humanity given recent developments in AI really
683.
by oytis 1y ago
Well, everyone on the internet is 10x productive with these tools, so it must be on you.
684.
by oytis 1y ago
But... Google's AI summaries are wrong like at least 50% of the time.
685.
by oytis 1y ago
It did already happen though. It's not just about buttons of course, there is a lot of lobbying, and if other people you are working with (e.g. other municipalities) stay on MS stack, there are going to be compatibility issues that wil
686.
by oytis 1y ago
Thank you, that's what I had in mind. The timeline is an amusing read, but as I read it it's really a mix of "buttons in the wrong places" and lobbying.
687.
by oytis 1y ago
True. It is already happened at least once in Germany that a municipality decided to ditch MS products in favour of open source, and went back after people started complaining that buttons are in the wrong places.
688.
by oytis 1y ago
Author's github history looks like an absolute coding machine
689.
by oytis 1y ago
> try-with-resources in Java, defer in Go, with in Python Or 'goto error8;' in C. Still RAII is much more convenient, especially for cases where you allocate a lot of interdependent resources at different time points. It keeps
690.
by oytis 1y ago
Repl is cool, but Python also has a repl, and has a much more intuitive programming model (and much fewer braces). Totally agree about the domain - no one is writing enterprise applications in C++ these days, luckily. It still does have its

More ›