First off, there are definitely cognates, because English borrows a ton of words from French (and modern French borrows some words from English), but in many cases the usage differs substantially.
For example, if I want to describe someone as very "sensitive" in French, I would probably say something like, "il est très sensible." Related word, but in English, "sensible" means something like "shows good judgment." To describe someone as sensible in French, you say, "il est sensé."
Secondly, the pronunciation differs a lot. The French "sensible" sounds something like "sonseebluh" to native English speakers. You have to have already tuned your ear to the sounds of French words to pick out most cognates.
It's absurd that this is even a question for some people. We may get AGI someday, but a vector database that generates streams of tokens based on tokens that it receives as input is never going to be it no matter how huge you make it. It lacks a world model, it doesn't have the capacity to update itself with new information, and - most importantly - it doesn't even do anything until it receives input.
And the assholes pushing this as potential AGI fucking know all that already. They know that without input, LLMs literally do nothing. They're not sitting there going, "I think therefore I am" while waiting around to hallucinate dangerous medical advice and fake case law at people. They are static and unchanging.
It's not a theory. A theory would have some evidence or reasoning behind it, even with the colloquial version of the word.
No, what's actually happening is a propaganda campaign. A really obvious one. And the same rich people proposing this "theory" are the ones directly and intentionally funding it.
The text on the whiteboard looks fake, her hand is weird, her arm is stumpy, the bar appears to be named "Bajmq" or something judging by the blurry text on the window, the background is blurred in ways that clearly disguise artifacts (and no actual camera of even moderate quality would do that these days)...
Edit: Oh, and don't forget the background people. Ugh.
It's slop. And the account posting it appears to be a bot.
As others have pointed out, the compute cost of inference is only one small part of the puzzle. All the frontier model providers - which OpenRouter gives you access to - are massively raising their prices in desperate bids to recoup the cost of model generation in the first place.
There's really no hiding from the token apocalypse unless you're running a model on your own hardware.
The unique thing about GitHub Copilot (and all the other vibe-coding tools) is that they're speed-running the playbook because this shit is not profitable. It can't be. Their costs scale up with usage, unlike every other business that can take advantage of economies of scale, so they've skipped the slow, steady enshittification phase and jumped directly into the "squeeze blood from this stone to keep the scam going a little longer" phase.
Out of curiosity, is his work on any of the "frontier" models? Or is he developing the small, open-weight models you can run on sufficiently beefy personal systems? There's a world of difference there in terms of ethics.
I work with a few AI engineers myself, and I don't think I know anyone who hates companies like OpenAI and Anthropic more than they do.
I'm just holding out hope that my company still has enough sanity left to keep from firing the team responsible for its (incredibly busy, active, and above all publicly-facing) developer portal in favor of having a vibe-coder destroy maintain it. At least long enough for the AI bubble to finally burst.
But LLMs are incapable of acquiring or understanding information.