<It is intentional.
I've said it a hundred times, M$ end game is total SaaS.>
Shite as a service?
28 publicly visible posts • joined 21 Dec 2024
<Meanwhile, the actual use case for AI is going great guns - research.
LLM's are cutting the time it takes to do deep research by 50% or more, which is impressive - this is where it is useful tech.>
I'm not sure where you're getting that it's so great for research, since almost everything I've read about Google's or Bing's "AI" assistants is that they make stuff up, conflate events and people, and assign the same level of importance to a white paper from a corporation and a Reddit response. These are not the kinds of things I look for in research assistance.
<No one owes you or your family anything; nor is it the local governments responsibility to support you during trying times like this! Sink or swim, it’s your choice! The City and County, along with power providers or any other service owes you NOTHING! I’m sick and tired of people looking for a damn hand out!>
It seems to sail right over Republicans' heads that people looking for a "hand out" are people who have paid their taxes and their power bills, and do indeed have a right to expect what they've paid for to be there when they need it. They get angry at "entitlements," but they're called that because people are rightfully entitled to what they pay for. I don't understand where this is coming from. Well, hate, I guess, against anyone who - uh - wants anything from anyone for any reason?
"For example, Intel's AI would have to be trained on everything Intel knows about designing and manufacturing integrated circuits in order for the AI to find a path out of Intel's current mess. Same with Boeing."
That might actually work. Even though hallucinations are endemic to LLMs because of the way they operate, they'd be much clearer to people who are familiar with the information it was trained on. There might even be fewer of them with a smaller training base.
"But it does offer a good insight to one possible use case. Flagging useful data hidden in a sea of data. As long as the sea of data is somewhat curated and validated I could see potential use cases."
The main (only?) thing LLMs (erroneously termed AI) are good for is pattern recognition.
"Why does advertising has to use personal information to be effective? Advertising has been around a lot longer than the ability to amass all this information, and I think the effect of personalized ads is overrated anyway."
DuckDuckGo is financing itself with context ads, which perform at least as well without requiring your information: https://duckduckgo.com/duckduckgo-help-pages/company/how-duckduckgo-makes-money
<They obviously don't care enough to do anything about it, won't you agree?>
They may not know that there's anything they can do about it. My husband is anti-tech, and although he does use the Internet (some), he wouldn't know about any of this stuff if it weren't for me. Even my dad, who is 89 but has enthusiastically used computers and the Internet for over 20 years, doesn't really understand that these issues exist or that there's anything he could do about them if he did know.
Those of us who know the problems and understand the various solutions are in a very small minority.
From Wikipedia:
"1978 Jonestown tragedy, where over 900 members of the Peoples Temple, led by Jim Jones, consumed a poisoned drink, often mistakenly referred to as Kool-Aid, during a mass suicide. In reality, the drink used was a cheaper product called Flavor Aid."
So, "drank the Kool-Ade" is short for "became a true believer even unto death."
"People who cannot drive may like the idea because it gives them independence in transportation that they now lack."
That's why I've been waiting for self-driving cars for years - I don't drive. My husband will take me places if I ask, and I have a ride service which has to be arranged ahead of time, but what I want to do is stop at interesting-looking shops that I hadn't known were there, or historical markers, or whatever, and there isn't really any way of doing that if you don't drive.
"Rubbish
Sorting. Now that is a good use for AI
Control robots that can sift through all the rubbish on a belt into the relevant categories for recycling or waste.
Will be better than what we have now (seeing some totally ignorant gits just chucking all their rubbish into the wrong bins)"
That's the first thing I've heard of where it might actually do some good for humans. That would be so much cheaper than people having to do a filthy job that no one wants, but which requires great accuracy. I hope someone who has some influence in that area picks up on this!
"Or...look at technology through the eyes of your 10 year old self...before your soul and passion shriveled up and died.
Tech has never been about ROI for me...for me, growing up, it was about making cool shit..."
Yeah, that's great for 10-year-olds, but grown-ups have to have some ROI or they won't be able to feed their own 10-year-olds.
"I know of one project where they are using it to read document from the 16 century and forward. Its not perfect yet but its getting there."
That's one of the things it's actually pretty good at. As far as I can tell, AI is good at spotting patterns, which has enabled it to turn up new materials and chemical formulations. It's apparently pretty good at translation, which I think is a kind of pattern detection.
It is not, and never will be, good at anything that requires reasoning. The ones that are out now claiming that they can do that, can't.
This is going to end up with NFTs, the Metaverse, and dozens of other things that looked bright and shiny but ended up wasteful and disappointing.
<Re: Delete them
If you read an article on the Internet that is freely available does that make your brain is now trained on that article and everything you write has to pay a royalty for?
If not why is it diffrent for a LLM?>
No one is using what my brain is "trained" with without my permission to make themselves a lot of money. If they want to, they'll have to talk to me about it. That's the difference.
>The Quarterlies are king, and we're actually an inconvenient stepping stone to that end
The requirement for quarterly reports has turned out to be one of the worst ideas capitalism has ever had. They not only need to raise productivity measureably every 3 months, but they have to raise it at a greater rate than the last quarter as well. This has been a big contributor to the rise and rule of the plutocrats.
<If Bing gives you the same answer as Marie Kondo when you ask it how to redecorate your living room, why does how it got to the answer matter?>
It matters to Marie Kondo, who is presumably paid a lot when she gives people this advice.
An evil not mentioned in the article is the blatant theft our imitation AI is based on. LLMs have no intelligence and are incapable of developing it, but can be used to steal intellectual property and anything else its owners want.
> The word "douche" is simply French for "shower", the kind you stand under in the bathroom to wash yourself.
> When and how did it become an insult?
I don't know where you are, but in the US the word is used almost exclusively for a feminine hygiene product. Anything used by women is automatically an insult when applied to men.
"They have machines doing that job already."
My husband went to McDonald's and was making his order. The sweet young thing behind the counter asked him if he wanted to try their cool new kiosk. He said, no, he doesn't use those because they replace the jobs people are doing. Her eyes widened and she said, "Do you think we should tell Corporate?"
(Easiest way to measure how far along it is, is probably to check how many rights to build ports along a new Northern Silk Road are being bought up all on the quiet by China.)
I've been wondering if Musk et al. have already been buying land around Hudson Bay, so that as the world heats up and the rest of the planet becomes to hot to live on, they'll have comfortable gated communities already set up.