112
current state of the AI bubble: the ceos are doing a vaguepost "hold me back bro" PR tour, Nate Silver is citing LessWrong, and the straits have somehow closed even further since we last checked (thelemmy.club)
submitted 2 days ago by to c/slop@hexbear.net
32 comments fedilink hide all child comments
 

yeah sure I'll believe people with immeasurable wealth and power who are actively accelerating climate change are suddenly "very concerned" about "existential risk" from the product they have staked their entire fortunes on the efficacy of.

billionaire-tears guys... grok is too powerful... grok is too world-changing...

you are viewing a single comment's thread
view the rest of the comments
[–] 52 points 2 days ago (17 children)

I’m really not sure what to make of this sudden wave of AI doomsday talk. It’s hard to tell who’s making a “ChatGPT will go Skynet any day now” argument versus who’s making “defense contractor CEOs will let AI make mission critical decisions unchecked and it’ll AI hallucinate an order to carpet bomb the eastern seaboard” arguments. Which I suspect might be the point, they’re trying to muddy the waters when it comes to AI criticism.

  • permalink
  • fedilink
  • source
  • hideshow 17 child comments
    •
  • [–] 57 points 2 days ago (8 children)

    They're trying to stall for more funding before the bubble pops

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 8 child comments
    •
  • [–] 26 points 2 days ago (7 children)

    It's taking an annoyingly long time for this bubble to pop! Hurry up already...

    I assume it's because it's the only thing propping up the US economy??

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 7 child comments
    •
  • [–] 19 points 2 days ago (6 children)

    Unlike the other tech bubbles, LLMs are a real technology with actual use value. They're overhyped, but not worthless. This gives the bubble room to grow, unlike previous tech bubbles which were hyping up worthless crap.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 6 child comments
    •
  • [–] 6 points 2 days ago (1 child)

    actual use value

    Citation needed.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 1 child comment
    •
  • [–] 4 points 2 days ago* (last edited 2 days ago)

    TL;DR: IAMAE and can't find a decent source because the internet is shit, but a good LLM won't modify your source data source and won't hallucinate an answer for you, it will give you the prompt. The bubble technology and how most people are using LLMs, and how capital wants you to use LLMs, is backwards. It's like going to a methed up doctor with a diagnosis and asking the doctor to prove your diagnosis to you, instead of the doctor soberly observing your health and telling you what if anything is wrong based on what they actually observe.

    Talk to anyone who worked with LLMs before ChatGPT was a household name and didn't work for OpenAI or Anthropic. like before 2019. This person will likely sigh deeply in anguish when you bring the subject up, because their whole career and practice has been bastardized. See https://www.smithsonianmag.com/smart-news/researchers-translate-bat-talk-and-they-argue-lot-180961564/ from 2016.

    The first use case for LLMs was basically analyzing large sets of structured (like a table with headers and consistent values) data. Like audio recordings of bats making noises. Or cancer biopsy results. With the right training and data, a LLM can identify patterns that a human likely couldn't based on looking at characteristics of the data that just wouldn't be apparent or the data set too large for a human to comprehend.

    The second use case for LLMs was analyzing unstructured data, like all the Discord messages with freeform text. Or pictures of fruit on a conveyor belt. A trained LLM could tell you if people were talking about agitprop or movie reviews, or if a piece of fruit is above or below a quality threshold because of it's size, coloring, shape, etc.

  • permalink
  • fedilink
  • source
  • parent
  • [–] 2 points 2 days ago (2 children)

    What other tech bubbles are you referring to that were less useful than LLMs?

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •
  • [–] 4 points 2 days ago (1 child)

    Guessing crypto and the metaverse. Which all blew up and deflated in ye olden times of 5 years ago

    dennis-stare

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 1 child comment
    •
  • [–] 7 points 2 days ago

    Yeah, and NFTs too. NFTs and metaverse are truly fundamentally useless. Crypto definitely has SOME use as it's become a major tool for international trade and can totally play a role in future currency structures. But obviously massively overhyped and primarily a speculative asset. But all of those are the bubbles of the last decade or so, after American tech plateaued. There were plenty of bubbles beforehand that produced genuinely useful and productive new technology. LLMs could have been a smaller one of those with the appropriate level of investment. The problem here is not really the utility of LLMs (which yeah, is real but not earth-shaking) and instead the MASSIVELY disproportionate capital investment that has upended the entire economy for no return to come.

  • permalink
  • fedilink
  • source
  • parent
  • [–] 12 points 2 days ago* (1 child)

    They literally used the AI to do target selection in Iran and apparently moved half the vetting team to other projects, according to some articles that came out after the first wave of attacks. So while the AI wasn't making the final decision, they used it to accelerate the target selection process, but that is it literally making the first initial decisions.

    They haven't actually admitted to it yet, but it is already being alleged, and I for one absolutely believe them because that is exactly the kind of dumb fucking shit they would do.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 1 child comment
    •
  • [–] 12 points 2 days ago

    They're using AI for the purpose of negating blame, not for any special advanced targeting data. They knew the girls' school they blew up was a school, they wanted to blow it up (killing officers' children is their intent) and used AI to launder the culpability.

  • permalink
  • fedilink
  • source
  • parent
  • [–] 34 points 2 days ago

    It's probably a way to poison the well and justify first strikes on China data centers. That only our glorious capitalist system can contain General AI for good, and China's inhuman communist system would only replicate our technology for Evil™️. When the 🔵Good President is elected we will finally get the regulations that the 🔴Bad President wouldn't give us because he was a ⚒️communist doing 🏦Corporatism instead of well regulated 😍Capitalism.

  • permalink
  • fedilink
  • source
  • parent
  • [–] 25 points 2 days ago

    I’m seeing it happen already, people responding to legitimate criticism by claiming it’s all tied into this skynet AI doom nonsense that’s apparently being pushed by Russia / China. Soon it’ll be the same as “Oh you believe in thing the CIA admitted to decades ago? Do you also think the moon landing was fake or the earth is flat too?”

  • permalink
  • fedilink
  • source
  • parent
  • [–] 23 points 2 days ago (1 child)

    The Millennium Prize Problems held significant symbolic value for the type of people who've been following AI doomsday scenarios since before LLM hype, so the recent thing where OpenAI might've solved the Navier Stokes continuity/smoothness problem has them all completely freaking out.

    It's a genuine reaction.

    Choosing to funnel money into those problems specifically was a deliberate marketing decision to cash in on that predictable reaction, though.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 1 child comment
    •
  • [–] 15 points 2 days ago*

    I was blown away by the possibility a LLM had made progress on the Navier Stokes equations, so I looked into it and

    OpenAI's solution has yet to be verified independently or publicly accepted by The Clay Mathematics Institute, a maths organisation based in the US which runs the Millennium Prize that offers big money to the first to solve certain mathematical conundrums.

    But the company's claim has already sparked a backlash.

    Tristan Buckmaster, a mathematics professor at New York University, said on Tuesday that he and Levent Alpöge, a mathematician working for OpenAI rival Anthropic, had also been working on solving the problem.

    The duo had been using OpenAI's tool Codex in their work. But Buckmaster said that on 3 September, he found out that "information about our progress had been passed to OpenAI".

    Buckmaster's statement came the same day, but hours before, OpenAI published its Navier–Stokes work. He claimed that OpenAI did not begin working on the Navier–Stokes equations until "after information about our work had reached OpenAI". He included text from emails exchanged with OpenAI over the work and his questions of the company's timing and methods.

    -- OpenAI says it cracked 90-year-old maths problem in 88 hour, BBC

    Lol, lmao.

    Breaking new grounds in plagiarism.

  • permalink
  • fedilink
  • source
  • parent
  • [–] 17 points 2 days ago

    I could see them pushing for some quick and industry-friendly regulation from Trump before he dies, mid-terms get blown out, or he leaves in 2028. Get the government to buy shares to keep this game going a little longer. If they're on the verge of collapse, then the government consolidating them out of national security and the good of mankind is a way out of it. Trump could just basically close down xAI, Anthropic, and OpenAI by buying them and folding it into the intelligence apparatus. It wouldn't be that hard considering they all do contract work anyways. There will be no stocks to tank. Shareholders get their money. Tax payers get another big industry bailout dressed up in reddit sci-fi dork kayfabe.

  • permalink
  • fedilink
  • source
  • parent