in theory several GBs .bin may be useful chatbot if it's not transformers even better. Not too easy to code but...for me as a user it's preferable over SaaS right now.
This is also an interesting case against LLMs/chatbots. If it was 1903 chatbots would've repeated the same that airplanes are impossible or that time is constant, there is no attraction between diamagnetic materials (casimir effect) or anything experimentally proven later. I'm not saying one should pursue claims that may always remain pseudoscience such as magnet motors but it helps to always doubt common academic consensus at the time and rather: experiment.
she looks gemini generated?? either musk made the profile "so cool, legalize comedy" or false flag from altman? 20% chance for a legit retarded doomer.
it just clicks with me, i like the syntax of the language or i believe it may solve important problems (e.q. AI helping with cancer cures, helping engineers design propulsion, improve loneliness etc). Existing documentation was important too back in the days before 2023 and the AI boom, for example asking question about c# on stackoverflow is one thing but asking about the Ring programming language totally different. Finally: versatility of language, i tend to avoid limited purpose languages like say perl or ruby which may be amazing idk but lack of gui tools, ai ecosystem, gaming development, app development etc i'd rather avoid them.
you can't seriously convince anyone over 15 they didn't anticipate this 'a priori', all Amodei, Askell, Cherny etc seem like very smart people...that's just hype for enterprise customers to pay more $$$ or for their upcoming ipo stock.
hmmm 10x, i added this line on top in the comment on top inside the file:
"PolyForm Noncommercial License 1.0.0"
(free for non-commercial)
^ I'm not sure it's the best license though, actually gpt 5.4 says it's a bad idea, but after a long puzzling wheather to make it GPL free for everything or refrain from posting on github at all i picked the middleground...i may change it later.
either you program AI itself or hardware/robotics: arduino/rasp pi.
I have a feeling the next "big thing" will be space but that's a bit premature, make it the mid-30s or 40s where the space-bubble starts.