LLM Hype is Already Descending, Apple Stopped Investing in the Money Furnace
Wall Street is a perverse force in the technology market, incentivising the most harmful (and mostly useless) things
THERE were many articles this past week bearing a false "valuation" of 'open' 'AI' - a company that never made a dime and loses money like it grows on trees (because Microsoft and NVIDIA provide de facto loans or help cushion massive - and fast-growing - debt). Some of the most senior roles have been abandoned/vacated lately. Key people are constantly fleeing (co-founder, CTO and so on). The company desperately needed to change the subject - and FAST.
This article isn't about finances (we covered this in IRC last night) and the gist of the story is, NVIDIA relies on the "AI" bubble to needlessly sell more and more useless chips (GPUs with "AI" branding/marketing); Microsoft bamboozles its investors with vapourware and false promises. So those companies would do anything they can to delay the bubble's implosion. This article mostly deals with the problems caused by the "AI" hype (usually not AI; LLMs lack actual comprehension, so they don't "lie", they just simply fail to function as advertised).
The hype is ruining companies, ruining technology, and this generally harms society the longer it goes on for. Some companies just go broke chasing a dud. It even ruins some formerly-'FOSS' companies.
Mozilla's cargo cultists have hopped onto the same bandwagon for no reason other than to mask the company's lack of real direction; buzzwords are so much easier and cheaper to name-drop than a substantiated strategy. Here we see a person (who used to work for Microsoft; the person talked about, not the author, who was a good journalist for many years) trying to tilt Creative Commons (CC) astray, or make it more plagiarism-friendly. Those people at Mozilla seem rather sceptical of at least some of the hype, recognising the limitation of "fair use" defence when LLMs are flagrant plagiarism and barely even work (even with the unwarranted plagiarism).
"The overall 'alignment chart' seems slanted towards the underlying presumption that some kind of plagiarism is desirable and acceptable," an associate of ours remarked. "The presentation there is thus kind of a false dilemma."
In Mozilla's case, even though there's no smoking gun/trigger (well, not yet), many people have taken it online and condemned the company for neglecting Firefox users to focus on advertisers and plagiarisers instead. People who use Firefox or Gecko-based browsers (siblings or cousins of Firefox) are struggling to access all sorts of Web sites, but instead of Mozilla' staff listening to them and fixing the issues it goes amok, seeking buzzwords and "monetisation" opportunities. This, in turn, drives away more users and it's making sites (or webmasters) even less eager to maintain Firefox support. By chasing users away Mozilla makes things worse. We lose the collective "strength in numbers".
This hopefully shows how the hype is misdirecting funds from useful technology to corrosive nonsense that burns the planet and delivers nothing of use (with solid use cases). Mozilla's blog talks a lot about "AI"; the company looks for money in the wrong places and alienates what it used to value: Firefox users (not advertisers, which Mozilla helped users block). █
"OpenAI announced this week that it has raised $6.6 billion in new funding and that the company is now valued at $157 billion overall. This is quite a feat for an organization that reportedly burns through $7 billion a year—far more cash than it brings in—but it makes sense when you realize that OpenAI’s primary product isn’t technology. It’s stories." - The Atlantic