The AI Bubble just popped

The Stock Market has had an AI bubble fueling its rise for almost 3 years now.
At some point, investors will say "enough!" and demand to see a return on their investment
(Not unlike the Dot Com bubble in Q1 2000.)
 

From the article:

“Like all other Chinese AI models, DeepSeek self-censors on topics deemed sensitive in China. It deflects queries about the 1989 Tiananmen Square protests or geopolitically fraught questions such as the possibility of China invading Taiwan. In tests, the DeepSeek bot is capable of giving detailed responses about political figures like Indian Prime Minister Narendra Modi, but declines to do so about Chinese President Xi Jinping.”

😳 😑 🤬

That ^ alone should result in a ban of the app in the United States; full stop. (See also: Tik Tok)

🇺🇸
 
A fake diamond is just as shiny as a real diamond and costs a fuckton less???

🤔
Both companies, DeepSeek and OpenAI, offer real, generative AI and reasoning large language models. DeepSeek is releasing a competitive SOTA image generating model, too.
 
Both companies, DeepSeek and OpenAI, offer real, generative AI and reasoning large language models. DeepSeek is releasing a competitive SOTA image generating model, too.

My point was they are both doing basically the same thing, but one is a fuckton cheaper…

Something is going on…

🤔

Side note:

I personally believe all AI is being overhyped and will ultimately under-deliver…or worse…

Sound familiar???

😑
 
My point was they are both doing basically the same thing, but one is a fuckton cheaper…

Something is going on…

🤔

Side note:

I personally believe all AI is being overhyped and will ultimately under-deliver…or worse…

Sound familiar???

😑
I think it's fashionable to dump on AI due to the amount of AI slop, grifters, hype, and misinformation. But I cannot agree that all AI is being overhyped. AI has already produced scientific breakthroughs, and is reshaping science, engineering, and technology.
 
Off-topic: PBS Documentary on the flamboyant and controversial US Rep. Adam Clayton Powell, Jr., (D-NY State 22nd, 18th and 16th): Powell knew how to push those trigger buttons back in the day.
 
I think it's fashionable to dump on AI due to the amount of AI slop, grifters, hype, and misinformation. But I cannot agree that all AI is being overhyped. AI has already produced scientific breakthroughs, and is reshaping science, engineering, and technology.

The known and potential cost(s) must be factored in.
 
I think it's fashionable to dump on AI due to the amount of AI slop, grifters, hype, and misinformation. But I cannot agree that all AI is being overhyped. AI has already produced scientific breakthroughs, and is reshaping science, engineering, and technology.

Where AI has real business application is intelligent pattern analysis in boring repetitive jobs, making it more accurate and a hell of a lot faster than humans. I do think it has genuine application in tech (it's already showing its usefulness in coding and medicine for just two examples). Sorting and classifying data more intelligently and flexibly than a straightforward database (Google has been doing it for years with best-guess fuzzy logic giving you meaningful answers from misspelt search terms for instance. The competition hasn't caught up with this yet; other search engines still look last-century in their rigidity).

Where it has (IMHO) zero use is in generating fake art images and craploads of inaccurately scraped and regurgitated text to flood Facebook in shite. Or plagiarising academic research for a lazy pass for college students.

I judge the online content I read based on its level of opinionatedness, or if it isn't opinionated, at least its human-feeling vibe. AI engines have been (so far) quite carefully engineered not to be opinionated, so there's no point asking an AI whether Fender or Gibson make better guitars, or Nikon or Canon better cameras. It will simply not give you the info you seek, and usually won't even give you an accurately targeted both-sides argument explaining the merits of either in a way that reflects real world experience.

The day I see a genuinely opinionated, convincing and believable AI-generated response that looks human is when I'll start worrying. It will probably happen, and that'll be the inflection point at which we recognise it is genuinely thinking and not just parroting. And that should scare us, because if machine intelligence can really think, it can rationally decide that we're too messy to be allowed to carry on existing.
 
This isn't a good thread. It's short-sighted.and ignorant of AI.
Negative Ghost Rider, the pattern is full.

Wall Street has lumbered from one technological marvel to the next since 2007. It's looking for the next cash cow, the next iPhone. Many have been very profitable, some have not. It looked as if self-driving cars would be Teh Next Big Thing, but sharp-elbowed operators have pushed self-driving cars out of the spotlight and focused on Artificial Intelligence. AI requires a shit-ton of resources to operate, and it's becoming pretty apparent that their is no clear path to profitability in the near to medium term. It's vaporware. The business model is flawed.

AI is probably the future....just not a future that you or I will live to see.
 
The internet giants are funding new nuke plants to run their LLMs. Skynet is a moot point when we so casually nuke ourselves. There will be at least more nuclear waste to dump somewhere, probably the southwest desert. With declining standards of competence and quality control, we may also have accidents at the plants, train derailments, truckers following google maps into lakes, etc.
 
Back
Top