Nvidia support is still poor (at least on latest cards), I'm forced to use X or I get tons of glitches.
I need the proprietary drivers for machine learning.
Not that I mind particularly, X is fine and everything works as expected.
I can even have the integrated (amd gpu) take care of the desktop while the Nvidia gpu does machine learning (I need all the vram I can get, the desktop alone would be 100-150mb of vram) - then I start a game on steam and the nvidia gpu gets used.
Funnily enough I had wayland enabled by default when I installed the system and I didn't understand why I was getting random freeze and artifacts for weeks. Then I realized I was not using X11
Your example is extreme, of course kenyan labelers contribute a minuscule fraction of value and they would be easily replaced by other workers if needed. Replacing the engineers is harder.
Labelling data is easier (and therefore cheaper) than doing machine learning.
OpenAI often argues for socialism-light, aka today's social democracies that leech on the middle class and favor politicians and the rich elite class.
It doesn't argue for full blown socialism with equal pay for any job. I think the historical proof is just too large to argue otherwise, which is why socialist politicians are going for the UBI alternative.
Same crap, but hey it will totally work this time.
This is the same axomatic ideology that GPT responded with. Miniscule fraction of what value? What do you think would happen to openai's valuation if their models would accidently advocate for pedophilia or building bombs? I mean this very topic we comment on just vaporized the market value of openai to single digit percentages. Openai raised nearly $18B, while DeepSeek-R1 cost $6M, works just as fine and is open. It has just been empirically proven that the market value of the developers working on GPT is actually dramatically lower than the payment they got.
Western AI models seem balanced if you are team democrats.
For anyone else they're completely unbalanced.
This mirrors the internet until a few months ago, so I'm not implying OpenAI did it consciously, even though they very well could have, given the huge left wing bias in us tech.
more literate voters -> more words -> word frequency patterns contain ideas that the model then knows.
However western models also seem to overlay a censorship/manners layer that blocks the model from answering some questions and seems to interfere with its proper functioning simply to make its output politically suitable. One example is to ask for a c program that will crash the linux kernel.
Chinese models get a lot of hype online, they cheat on benchmarks by using benchmark data in training, they definitely train on other models outputs that forbid training and in normal use their performance seem way below OpenAI and Anthropic.
The CCP set a goal and their AI engineer will do anything they can to reach it, but the end product doesn't look impressive enough.
I've considered homeschooling but ultimately decided against it because:
A - It's illegal where I live, you have to jump hoops to do it
B - I think an extra opportunity for socialising with kids is worth a bit of pain - we wouldn't have the energy for organising alternatives and also take care of the kids education
C - By sending kids to a nice private school you get rid of a lot of problems of public school: unmotivated / openly hostile and punishing teachers, classmates with bad behaviour disrupting class, immigrant classmates who don't speak the language and / or create gang of people from the same country to gang up on kids (exactly like in prison)
When I went to school things weren't as bad as today and kids were not getting stabbed in public school, still it felt like a prison because of the slow learning pace and because everyone learns at a different pace and wants to learn about different things. School is simply the wrong idea for the majority of boys, it's just a silly machine that print mindless employees.
The strongest reason for not sending them to school is the latest EU mandatory gender theory / sexual education propaganda being taught to kids in school since last year.
Ultimately I decided that years of socialisation with peers trumps a few lessons about political BS; I'm confident I can teach them to distrust authority and teach them that they cannot trust blindly everything they hear in school or on the newspaper.
My friends who pursued illegal homeschooling are quite happy, they even found a teacher who is teaching kids illegally in someone's home, and by grouping the kids together across multiple families they have a soft school experience.
This strategy works incredibly well and it's a continuation of their free dns proxy / caching service. It's a no brainer: the quality of the free services is unbeatable.
At the same time, everytime you need to buy something, you'll think "should I add a new cloud service or just buy Cloudflare?"
I don't like their almost monopoly-position but it's so good I use Cloudflare for all my projects and I keep recommending Cloudflare to all my clients.
Just a buzzword for investors given we peaked with language models.
Chaining different prompts can be useful: calling that agents is purely marketing: these models are pretty dumb and don't have agency. I'd stay away from related frameworks
Copy all photos / videos periodically to my linux server where everything important lives and gets backed up. Everything else gets nuked.
If I change phone I log into Google, start from scratch and download the apps I need / log in wherever I need the first time I need to.
I'm not very happy with using Google / Gmail for contacts and as a login manager, I've been postponing de-googling for 5+ years. The only promise I made to myself is that I'm not giving them a cent (and I keep living with 3% free space).
Not that I mind particularly, X is fine and everything works as expected. I can even have the integrated (amd gpu) take care of the desktop while the Nvidia gpu does machine learning (I need all the vram I can get, the desktop alone would be 100-150mb of vram) - then I start a game on steam and the nvidia gpu gets used.
Funnily enough I had wayland enabled by default when I installed the system and I didn't understand why I was getting random freeze and artifacts for weeks. Then I realized I was not using X11
reply