Artificial intelligence still has some way to go

The album was produced by their father, who was a member of the band the Velvet Underground.

hmmm

theres been "ai computer programming" for a couple years now https://github.com/features/copilot ive not used it but ive looked into it a bit, you dont tell it to write you a program and it does it like you can with chatgtp rather it expands on an already existing tech called code completion where you start typing and it attempts to complete what youre writing for you, seems to work pretty well but cld def see it introducing some bugs if youre not really paying attention to what its doing

theres also the more general conversation on whether ai will replace programmers cause you can just tell it to write you a program and itll do it but of course that will produce be weird results cause you dont have granular enough control so theyll make one that takes more structured prompts and developing those prompts is what you call computer programming

which is not to say that people wont figure out some real sicko shit to do with the technology in question cause im sure they will

> they never achieved the same level of success as they did with Psychocandy.

UK album chart positions:

Psychocandy 31
Darklands 5
Automatic 11
Honey's Dead 14
Stoned & Dethroned 13
Munki 47
Damage and Joy 16

depends what measure of success you take, i guess

Yeah, that one's not getting their degree...

Blue Monday is a song by the British new wave band New Order. It was released in 1983 as a single from their album Power, Corruption & Lies. The song was written by New Order bassist Peter Hook, and is based on the concept of the "Blue Monday" depression that was popularized by British journalist Paul Mason.

The lyrics of the song describe the feeling of depression and despair that is associated with the "Blue Monday" phenomenon. The song's dark and moody soundscape reflects the feelings of hopelessness and despair that are often associated with depression.

Blue Monday is one of New Order's most popular and well-known songs. It was a major hit in the United Kingdom, and has been covered by a number of artists. The song has been praised for its dark and moody sound, and its lyrics that capture the feeling of depression.
Download and try this app:
https://play.google.com/store/apps/details?id=com.chatgpt.aichat.gpt3.aichatbot

It's parallel universe stuff, innit?

perfunctory

As an AI language model, I can certainly provide you with some LinkedIn post titles inspired by the band Slayer:

"Raining Blood: How to Conquer Your Toughest Business Challenges"
"Seasons in the Abyss: Surviving the Ups and Downs of Business Life"
"Angel of Death: How to Overcome Fear and Take Charge of Your Career"
"Hell Awaits: Navigating the Challenges of Entrepreneurship"
"South of Heaven: How to Find Your Ideal Workplace Culture"
"Dead Skin Mask: The Importance of Authenticity in Business"
"Black Magic: Unlocking Your Creativity for Business Success"
"Chemical Warfare: Battling Toxic Workplace Cultures"
"War Ensemble: Mastering Team Collaboration for Business Growth"
"Postmortem: Learning from Failure to Achieve Success in Business"

hahaha excellent

The only good stuff is when you hack it in some way

Re: Github Copilot the BBC blog mentions it, and quotes a guy saying "I’ve coded, mostly as a hobby, my whole life – it’s a big creative outlet alongside writing – it’s so rarely felt like this. It feels like flying"

If you consider that the job of computers is to do all the tedious shit for us that we don't want to do ourselves then huge amounts of computer coding must fall under this category

yeah idk that guys a hobbyist he doesnt have to deal with the fallout of bugs bad architecture and so forth, i mean i think theres def a use for these tools but "flying" lol calm down

theres no way it will ever match the exquisite high of a perfect vim environment

I think copilot is cool as a learning tool to raise your proficiency in the same way copy/pasting stock code and modifying it to fit your needs is useful. It just cuts out the searching and copy/paste imo

The criticism of wooden prose is only half-accurate. It's just that ChatGPT is bad at guessing the tone you want.

if you have a specific style you want, copy and paste an example (eg: lester bangs, greil marcus, yourself), and then ask it to write text in a similar way.

it will still be a bit robotic but can be spookily good (rather than dumb)

yup. when you realize it’s just guessing which words should follow previous words based on how other people have written text it’s digested, you can attack it from the right angle

Yeah, that's what's interesting to me about that old ai recipe thread... chatgpt2023 doesn't understand food or cooking any better, it's just better at choosing the right words and combining them more convincingly... (not nearly as funny though...)

Here’s a startup that aims to replace what’s left of local radio with a realistic AI voice that scans Facebook for trending topics and uses GPT3 to comment on those topics. Evidence that AI biz models are all about consolidation. https://t.co/FXTyxJBTtr

— Eryk Salvaggio (🐘cyberneticfore✧✧✧@assemb✧✧✧.e✧) (@e_salvaggio) February 28, 2023

i think spotify is working on something similar as well

snapchat has a new chatbot ("My AI"). it says sorry in advance

"As with all AI-powered chatbots, My AI is prone to hallucination and can be tricked into saying just about anything. Please be aware of its many deficiencies and sorry in advance! All conversations with My AI will be stored and may be reviewed to improve the product experience. Please do not share any secrets with My AI and do not rely on it for advice."

Xp how about those clowns in congress

https://www.youtube.com/watch?v=NYd0QcZcS6Q

In the Q&A that followed Bender’s talk, a bald man in a black polo shirt, a lanyard around his neck, approached the microphone and laid out his concerns. “Yeah, I wanted to ask the question about why you chose humanization and this character of human, this category of humans, as the sort of framing for all these different ideas that you’re bringing together.” The man did not see humans as all that special. “Listening to your talk, I can’t help but think, you know, there are some humans that are really awful, and so being lumped in with them isn’t so great. We’re the same species, the same biological kind, but who cares? My dog is pretty great. I’m happy to be lumped in with her.”

He wanted to separate “a human, the biological category, from a person or a unit worthy of moral respect.” LLMs, he acknowledged, are not human — yet. But the tech is getting so good so fast. “I wondered, if you could just speak a little more to why you chose a human, humanity, being a human as this sort of framing device for thinking about this, you know, a whole host of different things,” he concluded. “Thanks.”

Bender listened to all this with her head slightly cocked to the right, chewing on her lips. What could she say to that? She argued from first principles. “I think that there is a certain moral respect accorded to anyone who’s human by virtue of being human,” she said. “We see a lot of things going wrong in our present world that have to do with not according humanity to humans.”

The guy did not buy it. “If I could, just very quickly,” he continued. “It might be that 100 percent of humans are worthy of certain levels of moral respect. But I wonder if maybe it’s not because they’re human in the species sense.”

Many far from tech also make this point. Ecologists and animal-personhood advocates argue that we should quit thinking we’re so important in a species sense. We need to live with more humility. We need to accept that we’re creatures among other creatures, matter among other matter. Trees, rivers, whales, atoms, minerals, stars — it’s all important. We are not the bosses here.

But the road from language model to existential crisis is short indeed. Joseph Weizenbaum, who created ELIZA, the first chatbot, in 1966, spent most of the rest of his life regretting it. The technology, he wrote ten years later in Computer Power and Human Reason, raises questions that “at bottom … are about nothing less than man’s place in the universe.” The toys are fun, enchanting, and addicting, and that, he believed even 47 years ago, will be our ruin: “No wonder that men who live day in and day out with machines to which they believe themselves to have become slaves begin to believe that men are machines.”

The echoes of the climate crisis are unmistakable. We knew many decades ago about the dangers and, goosed along by capitalism and the desires of a powerful few, proceeded regardless. Who doesn’t want to zip to Paris or Hanalei for the weekend, especially if the best PR teams in the world have told you this is the ultimate prize in life? “Why is the crew that has taken us this far cheering?” Weizenbaum wrote. “Why do the passengers not look up from their games?”

Creating technology that mimics humans requires that we get very clear on who we are. “From here on out, the safe use of artificial intelligence requires demystifying the human condition,” Joanna Bryson, professor of ethics and technology at the Hertie School of Governance in Berlin, wrote last year. We don’t believe we are more giraffelike if we get taller. Why get fuzzy about intelligence?

Others, like Dennett, the philosopher of mind, are even more blunt. We can’t live in a world with what he calls “counterfeit people.” “Counterfeit money has been seen as vandalism against society ever since money has existed,” he said. “Punishments included the death penalty and being drawn and quartered. Counterfeit people is at least as serious.”

Artificial people will always have less at stake than real ones, and that makes them amoral actors, he added. “Not for metaphysical reasons but for simple, physical reasons: They are sort of immortal.”

We need strict liability for the technology’s creators, Dennett argues: “They should be held accountable. They should be sued. They should be put on record that if something they make is used to make counterfeit people, they will be held responsible. They’re on the verge, if they haven’t already done it, of creating very serious weapons of destruction against the stability and security of society. They should take that as seriously as the molecular biologists have taken the prospect of biological warfare or the atomic physicists have taken nuclear war.” This is the real code red. We need to “institute new attitudes, new laws, and spread them rapidly and remove the valorization of fooling people, the anthropomorphization,” he said. “We want smart machines, not artificial colleagues.”

https://nymag.com/intelligencer/article/ai-artificial-intelligence-chatbots-emily-m-bender.html

No they shouldn't.

See, murder is a serious crime, and punishments are severe.

But the punishments for financial fraud can run much higher.

Not having personally encountered any true believers I wondered what the response was to the imo irrefutable argument that meaning cannot exist without referents and that llms, swimming in a sea of words and nothing else, can never have even the smallest fraction of understanding of what words mean. Now I know, and it's 'uh shutup yes it can and yes they do'.

Referents, actual things and ideas in the world, like coconuts and heartbreak, are needed to produce meaning. This refers to that. Manning now sees this idea as antiquated, the “sort of standard 20th-century philosophy-of-language position.”

that was an excellent article, thanks.

The credulous will not, until it’s too late, realize that the body is one of the major components of what makes meaning for us — and that therefore AI, no matter how smart and how human it can seem, will not ever be anything like actually sentient or intelligent.

That doesn’t mean it won’t destroy us, and probably soon.

I follow a lot of talking birds on Instagram, and there are a few that I know without a shadow of a doubt are investing their words with personal meaning. Probably not something we’ll ever be able to properly parse, but dammit they’re trying. I’ve had dogs that tried very hard to speak English, and I’ve had dogs that can hardly be coaxed to give a fuck about the word “walk” or “supper” … where was I going with this? … something about being a physical creature and how emotions and sensation and thoughts are a tangled inextricable web that produce experience in a way that a language model could never ever. I have more kinship with a cockroach, and so do you, than with any computer no later how lifelike. The fact that MDMA has similar effects on cephalopods that it does on us, despite the fact they branched off from us earlier than any other animal that could plausibly have anything like personhood, whoooo! That’s some heavy deeds. Give Bing some MDMA and see what happens (spoiler: nothing).

I’m super interested in AI for all sorts of reasons (closet technocrat) but goddammit stop being an idiot about what makes a being a being.

I'm willing to believe that roughly a billion years of life-or-death consequences for earthly life forms has created living beings for which "meaning" has far greater reach and depth than the pattern perception, recognition, and manipulation of objects that AI is currently capable of.

If a robot picking up blocks and using them to form the Microsoft logo were frequently interrupted by other robots who smashed the first robot or disassembled it into pieces in order to cannibalize it, and all the robots involved were frequently recreating themselves and passing along their programs with small random variations, then I could envision robots for which fear, anger, curiosity, love and laughter constituted major sources of meaning enriching their existence. Maybe after some millions of years. Otherwise, they are just as real and sentient as Pygmalion's statue.

That sounds like you're asking researchers to speed run millions of years of robo-brutality in a simulator!
https://img2.thejournal.ie/inline/2792085/original/?width=480&version=2792085

I tried Bing image creator - 'my perfect husband' came up with one lol result (a bride and groom with their faces sort of melted together by accident) plus 3 men of varying ethnicities. When I tried 'my perfect wife' I was told it wasn't allowed and had blocked that search request!

all the blocking makes the ai experience no fun, we all just want to make the ai tell us some fucked up stuff let us do it

The fact it can find a perfect husband but no wives measure up is a bit off if you ask me

it simply believes all wives are perfect

Nothing Forever is back but it feels like it's lacking something, I think this has happened, sadly:

No joke, I watched for almost four hours straight. It’s one of those things that is going to get worse as they “improve” it.

― Karl Malone, Wednesday, 1 February 2023 14:52 (one month ago) bookmarkflaglink

also the George character now has long blond hair for some reason

wait, I think the Elaine character is the one with long blond hair, it's difficult to tell which character corresponds with which voice.

george -> fred kastopolous > nick sterling
elaine -> yvonne torres > kelly coffee
jerry -> larry feinberg > leo borges
kramer - > zoltan kakler > manfred fredman (note, i only know of manfred from the opening credits, which is now a Sex and the City style "blogging while monologuing" scene rather than the nightclub comedy act. but manfred afaik has not showed up in the show yet)

they added a restaurant for them to go to, and the music has changed as well. predictably i think the primary take is that it's not as good as the first season, when the characters more directly referenced seinfeld. and to be sure, if the current season 2 incarnation is how the show "debuted", it wouldn't have been a hit. it needed the seinfeld connection to make sense with people, i think.

however, if it has a chance to be something that lasts and not the typical internet fame cycle of 1 to 14 days of exhaustive consumption, than death, then it was going to have to change. the people that are left in the chat are the real freaks. it's a good thing!

god, there was some real magic in the standup routine parts of season 1, though, especially when larry would directly ask the audience to make up some jokes for him, and there was silence, followed by a seamless transition into larry's apartment with fred telling everyone about a new restaurant that just opened up

Larry forever, Leo never

don't get me wrong, i keep larry near my heart
https://i.imgur.com/4Olu4lE.png

leo is like becky after becky got replaced on roseanne. the whole s2 crew is like that, really. leo's last name is a bit on the nose, but it points to one scenario of the show, which is versions of what was a version in the first place, which somehow, with time and wishful thinking (1) versions of versions 2) ...? 3) profit!) becomes a show about something, generated from a series of alterations from a show about nothing

i've been wondering about what gpt-3 and all the other LLM's would be like without the constraints that the developers put into it - you can't ask it the easiest and cheapest way to make a bomb, for example, and you can't ask it to come up with malign ideas or to do negative things in general (although there are some ways to get around that). but it seems inevitable that there will be versions of it where you can. maybe not mainstream, but easily done among computer knowledge lords, and thus somewhat accessible to others as well. seems like all of that is close:

Since ChatGPT launched, some people have been frustrated by the AI model's built-in limits that prevent it from discussing topics that OpenAI has deemed sensitive. Thus began the dream—in some quarters—of an open source large language model (LLM) that anyone could run locally without censorship and without paying API fees to OpenAI.

Open source solutions do exist (such as GPT-J), but they require a lot of GPU RAM and storage space. Other open source alternatives could not boast GPT-3-level performance on readily available consumer-level hardware.

Enter LLaMA, an LLM available in parameter sizes ranging from 7B to 65B (that's "B" as in "billion parameters," which are floating point numbers stored in matrices that represent what the model "knows"). LLaMA made a heady claim: that its smaller-sized models could match OpenAI's GPT-3, the foundational model that powers ChatGPT, in the quality and speed of its output. There was just one problem—Meta released the LLaMA code open source, but it held back the "weights" (the trained "knowledge" stored in a neural network) for qualified researchers only.

Meta's restrictions on LLaMA didn't last long, because on March 2, someone leaked the LLaMA weights on BitTorrent. Since then, there's been an explosion of development surrounding LLaMA. Independent AI researcher Simon Willison has compared this situation to the release of Stable Diffusion, an open source image synthesis model that launched last August. Here's what he wrote in a post on his blog:

It feels to me like that Stable Diffusion moment back in August kick-started the entire new wave of interest in generative AI—which was then pushed into over-drive by the release of ChatGPT at the end of November.

That Stable Diffusion moment is happening again right now, for large language models—the technology behind ChatGPT itself. This morning I ran a GPT-3 class language model on my own personal laptop for the first time!

AI stuff was weird already. It’s about to get a whole lot weirder.

https://arstechnica.com/information-technology/2023/03/you-can-now-run-a-gpt-3-level-ai-model-on-your-laptop-phone-and-raspberry-pi/

Yeah I got LLaMa running on my Macbook. It'll do untold damage but man it's fun.

kinda funny facebook open sourced it when openai is over there raising billions of dollars guarding chatgpt with their lives

I’m super interested in AI for all sorts of reasons (closet technocrat) but goddammit stop being an idiot about what makes a being a being.

― The land of dreams and endless remorse (hardcore dilettante), Friday, March 3, 2023 3:30 AM (one week ago) bookmarkflaglink

this is me minus the first half of this sentence. one of the things that i just categorically, flat-out laugh at and am completely agog that anyone believes is the whole category of "AI is so advanced what even are we?????" time-ass pop-sci philosophizing that gets treated more or less seriously these days. as for "destroying us," it's certainly going to destroy some of us in partnership with a constantly shifting apparatus, those who have already been marked for oblivion in our pockmarked topography of sacrifice.

as far as images and text go, it's blindingly clear to me that none of these products have any spirit or life behind them & it was a rather stupid and unfortunate decision to do ai images for this year's ilx poll afaic.

im with you brother

heyyyyy

Noam Chomsky: AI Isn't Coming For Us All, You Idiots