astromaniak

7 ms·

31.
by astromaniak 2y ago
Interpretation of "reasoning" looks to be a matter of opinion, like AI and AGI. If it's sequential application of some rules, then there is a lot of 'reasoning' software around. Almost all of it. There is no strict
32.
by astromaniak 2y ago
With GPT4/o there is a trick for math problems. You can ask it to write the python code. This solves for example famous problem of counting letters in string. Sure model can be trained to use python under the hood without being explici
33.
by astromaniak 2y ago
Just in case you are a pro developer, the whole thing is worth looking at: https://github.com/CppCon/CppCon2017/tree/master/Presentatio... and up
34.
by astromaniak 2y ago
Under the carpet deals would make it less transparent. It would be hard to detect that top manager sold all his stock to competitors and now is making decisions in their favor.
35.
by astromaniak 2y ago
> Warren Buffett proposed that the stock market should be open less frequently This will result in another market where deals will be made and then finalized on that 'official' when it opens. It's like with employee stock.
36.
by astromaniak 2y ago
Finally they realized why NVidia wins. It's the whole coherent ecosystem, not just hardware with minimalist support in form of buggy drivers.
37.
by astromaniak 2y ago
The difference has two components: genetic, personal. Genetic does not depend on self identification. Personal is very similar these days. Same families, same schools. As for trans. Humans, unlike most animals, cannot even have sex without
38.
by astromaniak 2y ago
it's very limiting probably too. likely they cannot have side contacts with competitors while having insider info in oai. also they likely cannot develop competing products themself. and obviously non disclosure. all this makes it a st
39.
by astromaniak 2y ago
There was a report about training small model on eventset completely generated by GPT4. Small stories using kids vocabulary. Probably (half) a year back. So it's possible. My idea was to mix in significant portion of generated texts o
40.
by astromaniak 2y ago
> So if Superintelligence depends on AGI, Not sure, is superintelligence generic, alive. Does it have motivation, goals? Sounds like a big calculator which he personally will control. Technically it's possible. LLMs aren't the
41.
by astromaniak 2y ago
> First, LLMs are not AGI. It's the most generic thing we have right now, right? > Never will be. If there is no other breakthrough anytime soon we can engineer AGI-like things around LLMs. I mean LLM trained to use different att
42.
by astromaniak 2y ago
I'm afraid this is going to be another Everyday Robots or Boston Dynamics. Google is ruled by managers, not visionaries. Strategy is 'try and see if it works in 4 years. if not cancel'. They followed it in many cases. So, De
43.
by astromaniak 2y ago
BTW, while EU laws protect personal data. From being exported, for example. Raw video and images should be ok to export and do actual recognition somewhere else, where laws are more permissive. No border crossing of personal data.
44.
by astromaniak 2y ago
I had even bigger "brother" for significant part of my life. That was more than enough.
45.
by astromaniak 2y ago
LLMs aren't humans. you can be pushy without being rude. In cases like this I simply ask for the full version. Usually ChatGPT produces it. GPT4o is more verbose, so this should be less of a problem.
46.
by astromaniak 2y ago
Will it be as good. I mean control, video recording and signal. Compatible with some HD goggles? I really doubt, but if you can explain...
47.
by astromaniak 2y ago
> Some Autel drone are made in the USA You mean are assembled in USA. Most soft and hardware is still made in China.
48.
by astromaniak 2y ago
That's only official part. Having cameras you can track a lot. 1. recognize people 2. track who is interacting with whom 3. track who personally pays attention to what ads on the screen 4. get some clues about health and wealth. perso
49.
by astromaniak 2y ago
> "use a range of advanced technologies across our stations to protect passengers" Total surveillance. Of course, that's for your own good. Do they guarantee that after 'smile detection' image is deleted. Or they
50.
by astromaniak 2y ago
it can pretend, which often looks indistinguishable. so that you can't say if it's real or not.
51.
by astromaniak 2y ago
just wonder, do you know any opens source LLM? that can be reproduced?
52.
by astromaniak 2y ago
good weights are the key, but it has python interpreter and bing search. and likely more in development.
53.
by astromaniak 2y ago
It can't be, it's too big to die.
54.
by astromaniak 2y ago
babies do a lot more. they have feelings. like pain, pleasure. and different verities of them. then they have a lot of things hardcoded. and they have control of their bodies and the environment. for example they quickly learn that crying h
55.
by astromaniak 2y ago
doing business is hard this days. you can't be just a rich a*hole while working with people. have to care about your image. and this is one of the ways of doing it. hanging out free stuff. sort of selfless donations. but in fact this r
56.
by astromaniak 2y ago
Exactly. and that's why it's valued at $3T++, about 10x of AMD and Intel put together.
57.
by astromaniak 2y ago
They have the right to have their opinion and build it into their products. But how about having other opinions and products too? Imagine in hospital, or care houses, patients will be more comfortable with human-like AI. Or AIG when it happ
58.
by astromaniak 2y ago
Easy enough, ChatGPT can write verification code along the main line. Just ask.
59.
by astromaniak 2y ago
No, they cannot work with large code base, not yet. And have very limited talent for logic and debugging. They may improve at some point, probably will be hooked up with external tools.
60.
by astromaniak 2y ago
It's probably easier to reinstall everything anew from time to time. Instead of fixing broken 18.04 just move to 22.04. Most tools should work, if you don't have huge codebase which requires old compiler... Conda.. it interfere wi

More ›