Artificial Intelligence
- Started
- Last post
- 1,639 Responses
- neverscared7
- tuché_niko
- touchéhydro74
- Twoshay_niko
- tooshiePonyBoy
- Too cheeseOBBTKN
- Exactly. Who’s reading all this AI written stuff? Bots scraping AI written blog articles for SEO ranking.shapesalad
- Who’s looking at all these AI generated images? Who has time for that.shapesalad
- Stanley Tucci. /endmonospaced
- uan3
I asked chatGPT to simulate a conversation between Simone Weil and Albert Camus...it started well, but when the simulation started to disagree I got this:
- got a second network error when asked again.
I hope I didn't break it asking stupid suff :-)uan - You killed AI!palimpsest
- i also kept breaking it trying to get it to give me the code for streaming a linkedin post feed to a webpagehans_glib
- hearing that llama-7b deals a lot better running this sort of thing, GPT4 cuts you off when it falls in a loop justkingsteven
- this means my plan to compile 'the digital anarchist cookbook' will take a lot more work.uan
- you broke it!dbloc
- got a second network error when asked again.
- oey_oey0
I just received the following warning in chat GPT:
Error in body stream
:-)
- Continuity0
"Elon Musk joins call for pause in creation of giant AI ‘digital minds’
More than 1,000 artificial intelligence experts urge delay until world can be confident ‘effects will be positive and risks manageable’"
“Recent months have seen AI labs locked in an out-of-control race to develop and deploy ever more powerful digital minds that no one – not even their creators – can understand, predict, or reliably control,” the letter says.
“Powerful AI systems should be developed only once we are confident that their effects will be positive and their risks will be manageable.”
- I'l come right out and say it: AI terrifies the shit out of me. We're playing with fire.Continuity
- *I'llContinuity
- Also: I may or may not have read and re-read Gibson's Sprawl trilogy one too many times.Continuity
- lol whatgrafician
- It needs to be stopped and full sceintific/economic impact assessment done. Because if we keep going ahead with AI, we will have huge unemployment.shapesalad
- which will lead to... violenceshapesalad
- until they can catch up and monetize the shitbezoar
- PhanLo7
tried some GEN-1
-
- input clip https://i.imgur.com/…PhanLo
- As often - bravo!Nairn
- I like that it recognised the streetlampPhanLo
- good stuffdbloc
- I'm so envious of the time you appear to have available to you for this sort of thing.
I mean this earnestly.
TEACH ME HOW.Nairn - i suspect first point would be "Leave London".Nairn
- I do the look at screen all day and then pack it up, go home unpack and then look at screen again method. The Gen-1 thing was so fast though, super easy.PhanLo
- I was waiting for the Runway text to video invite but I've just noticed your mockup. Great job!palimpsest
- What I find sad is that people don't understand the work behind the machine and the work being produced by the person. I love that you hand crafted the prompt!palimpsest
- I'd really like to build a small set of the buildings and eventually put 2d animated characters in the scene. Trying some photogrammetry today.PhanLo
- nice!!!utopian
- sted0
- NBQ002
/imagine: QBN website in Wes Anderson style, pastel colours, minimalistic design, not trending on art station, 8k
- https://i.imgur.com/…palimpsest
- https://i.imgur.com/…uan
- The gift that keeps on giving.palimpsest
- palimpsest-1
Honest question.
How many of you see our current AI technology the same as or close to HAL 9000 or Skynet?- Immediately? No, not close. Not as long as AI can't understand that 8 fingers on a human hand ain't right. In 5 or 10 years? Quite possibly.Continuity
- I would love to see it. But so far I don't see anything pointing in that direction. As I see it could come in 5 years or never.palimpsest
- Well, look. Humans are dumb as bricks, and they're always doing shit they shouldn't be doing:Continuity
- developing nuclear weapons, ruining the planet by burning hydrocarbons, letting romanian neckbeards on QBN. If so many people from the article I posted are ...Continuity
- ... concerned (escpecially a few experts, it would seem), then I think we should be, too.Continuity
- When our AI technology can say "I don't know"drgs
- I still haven't read the article:
https://www.qbn.com/…palimpsest - The question was prompted from a short interview I just heard. Where the specialist voices concerns over AI giving wrong information (fake news).palimpsest
- But also said that we were far (not in the works) of it being sentient which is how I understand current technology.palimpsest
- If the worry is fake news, to me it just sounds the same as the complaint that video games and music cause violence.palimpsest
- It seems to me that the fear of AI doesn't come from the unknown but from the known based in science fiction.palimpsest
- All we did in the past decade(s) is that we turned the toys from syfy movies into reality without changing the fundamentals of our thinking.sted
- So this fear is an expected consequence.sted
- I don't think that there aren't any dangers associated with AI but as I understand it nothing points to it becoming sentient.palimpsest
- Continuity went nuts over this shit :))))
chill bros there no AGI soon
QBN understanding of AI is shit at best
and you're calling me arrogantgrafician
- palimpsest0
"You raise a valid point that the potential risks and downsides associated with AI often stem from human decision-making and actions, rather than from any inherent flaws in the technology itself. Just as humans can misuse and abuse the vast amounts of information available on the internet, we can also use AI systems in ways that are harmful, biased, or unethical.
One of the challenges with AI is that it can often operate on a level that is beyond our conscious awareness, which can make it difficult to monitor and regulate its use effectively. In addition, the algorithms and models used in AI are often based on large datasets that may contain biases or inaccuracies, which can lead to unfair or harmful outcomes if those biases are not identified and corrected.
At the same time, I do believe that AI has the potential to be a powerful force for good, and that with careful oversight and responsible decision-making, we can develop AI systems that are fair, transparent, and beneficial for all. This will require ongoing collaboration and dialogue between experts in AI, policymakers, and the public, as well as a commitment to ethical and responsible use of this technology."
ChatGPT
- Do people get that these models model words in a sentence - barely?
That there's no "AI" present
Fucking chumpsgrafician
- Do people get that these models model words in a sentence - barely?
- palimpsest0
"As an artificial intelligence language model, I only respond to user inputs that are provided to me. I don't have any independent thoughts or actions of my own, and I don't have consciousness or any internal experiences. My responses are generated by processing the input text and providing the most appropriate response based on the patterns and correlations I have learned from the vast amount of data I was trained on.
My ability to understand and generate human language is the result of a complex combination of statistical learning, natural language processing algorithms, and deep learning techniques. While this can appear to be intelligent behavior, it is important to note that I do not have consciousness, emotions, or subjective experiences like humans do. Rather, my responses are generated by processing and analyzing the input text, without any understanding or awareness of the content or context beyond the language itself."
ChatGPT
- As far as I can count, ChatGPT said the same thing four different ways.Continuity
- Now, ask GPT if it's governed by Asimov's Three Laws of Robotics, plus Zeroth Law.Continuity
- ChatGPT has access to knowledge which is verbalizable. The vast majority of human skills, experiences, are exactly this -- difficult to put into words.drgs
- Like have a certain intuition about physics, how fluids behave etc. Yann LeCun has a lecture on how sensory grounding is needed for full meaning and undrstdngdrgs
- Thanks, I'll look for it. In the "converstation" I was having with it. I also asked it to describe a smell.palimpsest
- I then argued that the experience of how it got that info was irrelevant and we both could hold the same concept of smell.palimpsest
- But it's not really about the experience of the smell it's about how that information is connected and put into context with other information we humans hold.palimpsest
- While AI has a linear approach to accessing information. It doesn't make all the (unconscious) connections we make when accessing a piece of data.palimpsest
- Bald statements coming from Continuity lol
Bro these are language models theres no Asimov's Laws applied to this shit :))))
Fucking chumpsgrafician
- Morning_star2
Is our current AI technology the same as or close to HAL 9000 or Skynet?
"Fuck off Dave i'm wanking" - ChatGPT
- Continuity2
I'm not reassured.
- Delete System32 just in case.palimpsest
- Has anyone tried asking ChatGPT to delete itself?shapesalad
- "As an AI language model, I don't have the ability to delete myself on my own, since I exist as a program running on servers maintained by my creators at OpenAIpalimpsest
- However, my creators have put in place various measures to ensure that I operate safely and securely, and they have the ability to shut down or modify my codepalimpsest
- if necessary. Ultimately, though, the decision to do so would be up to them, not me."palimpsest
- I'm sure Continuity will sleep better tonight.palimpsest
- Cringe af
Who tf asks this question to an non-AI entity??!?! Imposibrruuuuuugrafician - Upvoting this, this is fucking epicuuuuugrafician
- Get a fucking hold of yourself!palimpsest
- palimpsest2
Me: "Do aesthetics always need to be about the retinal? What about the concept of an elderly obese woman sitting on the back of a Cadillac breastfeeding a Chihuahua?"
ChatGPT: "Aesthetics do not necessarily have to be solely about the visual or retinal experience. Aesthetics can also involve sensory experiences beyond sight, such as touch, taste, and sound, as well as concepts and ideas.
The example you provided of an elderly obese woman sitting on the back of a Cadillac breastfeeding a Chihuahua challenges traditional notions of beauty and aesthetics. However, it can still be considered aesthetically interesting and thought-provoking because it creates an unexpected and unconventional juxtaposition of elements that can evoke different emotions and interpretations.
Aesthetics is not limited to conventional beauty or visual appeal, but can encompass a wide range of experiences and interpretations that engage our senses and emotions. Ultimately, the interpretation of what is aesthetically pleasing or interesting is subjective and depends on individual preferences and cultural context."
- palimpsest0
ChatGPT: "The concept of an elderly obese woman sitting on the back of a Cadillac breastfeeding a Chihuahua could be considered camp, depending on how it is presented and interpreted. Camp is a cultural style and sensibility that often involves exaggeration, irony, and humor. It can involve an embrace of the artificial, the theatrical, and the kitsch, and a rejection of conventional notions of taste and propriety.
The concept of an elderly obese woman sitting on the back of a Cadillac breastfeeding a Chihuahua certainly involves exaggeration and irony, as well as an unexpected juxtaposition of elements. Depending on the presentation and context, it could be interpreted as campy.
However, whether or not this concept is camp depends on how it is presented, as well as the interpretation of the viewer. The concept could also be presented in a way that is not campy, but instead presents a serious social critique or explores a different aesthetic sensibility."
- Nairn5
The on-going argument I've been having with my best pal about all this - and I guess to oblivious arrogant pricks like grafician who think they're positioned to gatekeep this shit because they have no capacity for nuance or subtlety, so polarise their interpretation of others interest in the domain in their favour (and because they read two articles on the sibject on Hacker News yesterday and are now Expert-Level™) - is that this doesn't need to get anywhere near AGI levels to begin to cause profound societal trouble. Nowhere near. At all.
We're probably a [human] generation or three away from actual AGI as feared from sci-fi tropes, but all we need now are simulacra that can generally pass Turing tests or whatever contemporary equivalents could be mustered, and so can sow discord in their own way. As directed by whoever decides to.
Think of it as 'Russian political meddling on amphetamine and lightning'.
Never mind the slew of contemporarily somewhat useless from a productivity-persepective but essential from a societal-perspective jobs that could be nixed within the next couple of years, possibly literally.
English loom workers of yore at least had a generation to realise they needed to start breaking machines before they could re-skill. No such luxury today.
.
Which is to say, yes we get it graf - it's not actual fucking ai, it's clever algorithms refined on mind-bogglingly large datasets. it's just that everyone calls it AI. We're older than you - we also had people imoplying that fuzzy logic in washing machines in the 90s was 'intelligent'. Actually, your arrogance tells us more about your ignorance than anything else.
- Turn off the computer and non of this AI stuff exists. It's so nice to walk through nature and smell the plants hear the birds. Fuck AI + computer's.shapesalad
- 20 year career spent in front of a computer. Frankly and enough of it. Would love to hit 'shutdown' and never have to turn on a computer again.shapesalad
- Amen!palimpsest
- *had enough. Had enough of typing too...shapesalad
- Protect people, not jobs.palimpsest
- Amen @Nairn, to be clear.palimpsest
- To state the obvious, words mean wathever we agree them to mean. So AI does refer to the current technology we have at hand.No need to throw a tantrum about it;palimpsest
- ^ @graficianpalimpsest
- I understand the implications of this technology being a major tipping point in human history, but to suggest we just 'switch it off' is pretty lazyIanbolton
- lol Nairn
I just wish old people here just take a day or two and get to read and understand the tech so we can all be on the same page?
Not that hardgrafician
- neverscared1
Italian privacy regulator bans ChatGPT
Calls have grown to suspend new releases of popular AI tool.The Italian privacy regulator Friday ordered a ban on ChatGPT over alleged privacy violations.
The national data protection authority said it will immediately block and investigate OpenAI, the U.S. company behind the popular artificial intelligence tool, from processing the data of Italian users. The order is temporary until the company respects the EU's landmark privacy law, the General Data Protection Regulation (GDPR).
- Bans how? Italians can't access the site?drgs
- They don't ban it as such, just that in Italy they tie the AI's hands behind its back so it can't do any real harm.Nairn
- It's this like the EU not letting apps spy on people?
Thank god for Brexit!palimpsest
- sted3
"bro I was out with me friends having beers
while you guys here thinking about me "
#WoodstockAI in SF- https://i.imgur.com/…sted
- My dude!palimpsest
- where's the cat sted?grafician
