this post was submitted on 19 Oct 2023
540 points (96.6% liked)
Technology
59217 readers
3155 users here now
This is a most excellent place for technology news and articles.
Our Rules
- Follow the lemmy.world rules.
- Only tech related content.
- Be excellent to each another!
- Mod approved content bots can post up to 10 articles per day.
- Threads asking for personal tech support may be deleted.
- Politics threads may be removed.
- No memes allowed as posts, OK to post as comments.
- Only approved bots from the list below, to ask if your bot can be added please contact us.
- Check for duplicates before posting, duplicates may be removed
Approved Bots
founded 1 year ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
The thing with AI, is that it mostly only produces trash now.
But look back to 5 years ago, what were people saying about AI? Hell, many thought that the kind of art that AI can make today would be impossible for it to create! ..And then it suddenly did. We'll, it wasn't actually suddenly, and the people in the space probably saw it coming, but still.
The point is, we keep getting better at creating AIs that do stuff we thought were impossible a few years ago, stuff that we said would show true intelligence if an AI can do them. And yet, every time some new impressive AI gets developed, people say it sucks, is boring, is far from good enough, etc. While it slowly, every time, creeps on closer to us, replacing a few jobs here and there in the fringes. Sure, it's not true intelligence, and it still doesn't beat humans, but, it beats most, at demand, and what happens when inevitably better AIs get created?
Maybe we're in for another decades long AI winter.. or maybe we're not, and plenty more AI revolutions are just around the corner. I think AIs current capabilities are frighteningly good, and not something I expected to happen this soon. And the last decade or so has seen massive progress in this area, who's to say where the current path stops?
Nah, nah to all of it. LLM is a parlor trick and not a very good one. If we are ever able to make a general artificial intelligence, that's an entirely different story. But text prediction on steroids doesn't move the needle.
The best ones can literally write pretty good code, and explain any concept on the Internet to you that you ask them to. If you don't understand a specific thing about their explanation, they can add onto their explanation, and they can respond in the style you want (explain as if I'm ten, explain as if I'm an undergrad, etc).
I use it literally every day for work in a somewhat niche field. I don't really agree that it's a "parlor trick".
LLMs are awful for facts, because they don't understand what facts are. You should never rely on them if you require factual correctness.
They are OK for text summation, formatting and just making shit up. For summation a human with experience still produces nicer output, because they understand the content and don't just look at words. As for making shit up you will get the statistically most likely output, so it's usually trite and boring. I think the progress is amazing, but there are still so many problems to be solved.
Right now I use them for boiler plate stuff, like writing a text with some parameters and then I polish it. For code I find them quite useless, because with an IDE I can write boiler plate just as fast as when I polish the prompts until the LLM delivers useful stuff. And with the IDE I don't get references to methods or entire libraries that just don't exist.
It's actually great for dnd to produce NPC dialogue or names on the fly. We also tried using it to calculate area of effect spells, ie "how many average sized humans in armor with swords could fit in a circle with a diameter of 30ft." We were rolling with it before someone pointed out that it didn't calculate the area of a circle correctly, however it got the rest more or less accurate. So we don't use it for that anymore, and it's funny how what often appears to be the simplest component of a question is the thing it most often gets wrong.
People are also kind of shit at facts. There are so many facts, and many of them aren’t practical for every person who needs to assess a fact’s accuracy to do so. But it isn’t structurally impossible to mimic how humans learn how to gauge truthfulness, we just have to be prepared for the idea that it will be bound by the limitations of language, as well as the risk inherent in trusting data that it has not independently verified.
I use LLMs for having things explained to me, too.. but if you want to know how much salt to pour in that soup, try asking it about something niche and complicated you already know the answer to.
They can be useful in figuring out the correct terminology so that you can find the answer on your own, or for pointing some very very obvious mistakes in your understandings (but it will still miss most of them).
Please don't use those things as answer machines.
I'm going to use those things as answer machines and you can't stop me.
Jokes aside, I always validate what chatbots tell me, not even just important things. I use GPT-4 for work and 90% of the time it can show me how to use very specific functions in complex ways, but yesterday (for the first time in awhile) it made up a function that didn't exist. To its credit, I said, "Are you sure about [function]?" and it said, "I'm sorry, I got confused. That function doesn't exist. However, look into X, Y, Z for further resources" and I did and they were the correct things to look into.
If you press it the same way again ("are you sure the function doesn't exist?"), there is a high chance it will "rectify" its rectification.
Sam Altman (Creator of the freakish retina scanning based Worldcoin) would agree, it seems. The current path for LLMs and GPT seems to be in something of a bind, because to seriously improve upon what it currently does it needs to do something different, not more of the same. And figuring out something different could be very hard. https://www.wired.com/story/openai-ceo-sam-altman-the-age-of-giant-ai-models-is-already-over/
At least that's what I understand of it.
He's not saying "AI is done, there's nothing else to do, we've hit the limit", he's saying "bigger models don't necessarily yield better results like we had initially anticipated"
Sam recently went before congress and advocated for limiting model sizes as a means of regulation, because, at the time, he believed bigger would generally always mean better outputs. What we're seeing now is that if a model is too large it will have trouble producing truthful output, which is super important to us humans.
And honestly, I don't think anyone should be shocked by this. Our own human brains have different sections that control different aspects of our lives. Why would an AI brain be different?
Future of AI is definitely going towards Manager/Agent model. It allows for an AI to handle all the tasks without keeping it to one model or method. We’re already seeing this with ChatGPT using Mathematica for math questions. Soon we can see art AI using different models and methods based on text input.
I gather that this is partly because data sizes haven't been going up with model sizes. That is likely to change soon as synthetic data starts to overtake organic data in both quantity and quality.
In humans, abstract thinking developed hand in hand with language. So despite their limitations, I think that at least early AGI will include an LLM in some way.
I've been having a lot of vague thoughts about the unconscious bits of our brains and body, in regards to LLMs. The parts of our brains/neurons that started evolving back in simple animals as basically super primitive ways to process visual/audio/whatever input.
Our brains do a LOT of signal processing and filtering that never reaches conscious thought, that we can't even reach with our conscious thought if we tried, but which is necessary for our squishy body-things to take in input from our environment and turn it into something useful instead of drowning in a screeching eye-searing tangled mess of chaotic sensory input all the time.
LLMs strike me as that sort of low-level input processing, the pattern-recognition and filtering. I think true generalized AI would have to be built on pieces like this--probably a lot of them. Ways to pluck patterns out of complex but repeated input. Like, this stuff definitely isn't self-aware, but could eventually end up as some sort of processing library for something else far down the line.
Now might be a good time to pick up Peter Watts' sci-fi book Blindsight. He doesn't exactly write about AI in it, but he does write about a creature that responds to input but isn't exactly conscious like you or I.
This is what I meant.
I just got the EPUB, thanks. Looking forward to reading it.
FYI Blindsight is free for audible members.
Parlor trick is a perfect description.
People don't get that these things aren't anymore intelligent than their smartphones predicting the next word. The main difference is instead of a couple words it has thousands to choose from.
Half of the trick is how it uses the prompt to decided what words to start with.
That is not how it works. Your smartphone has all the dictionary available, same as LLM. It is simply something very different. People super confidently discussing about AI on lemmy are the real hallucinating parrots
Yes. I've used them. I have used it beyond the point of it hallucinating.
I am also a software engineer and have deeper understanding of how these systems work than your average user.
The software community tends to approach these things with more caution than the general population. The media overblows the capabilities of these systems.
A more concrete example is autonomous vehicles which were promised for decades and even now with a form of them on the road, they are still closer to remote controlled vehicles than the intelligent self contained systems we have been promised.
The difference between predictive text on a smart phone and predictive text of an LLM is my smart phone is predicting what I am likely to type next based on things i have typed in the past, while the LLM is predicting what comes next based on a larger body of work from source pulled from all across the internet. The LLM is then tuned by humans. This tuning step is under reported.
The LLM is unable to determine the truth of its own output. I would argue that is a key to claiming intelligence but determining what intelligence means is itself a philosophical question up for debate.
Yeah exactly and a great way to see this is by asking it to produce two viewpoints about the same subject, a negative and positive review of something you're familiar with is perfect. It produces this hilarious "critic" type jargon but you can tell it doesn't actually understand. Coincidentally, it's drawing from a lot of text where the original human author(s) might not understand either and are merely themselves re-producing a jargon-heavy text for an assignment by their employer or academic institution. If AI can so accurately replicate some academic paper that probably didn't need to be written for anything other than to meet publishing standards for tenured professors, then that's really a reflection on the source material. Since LLM can only create something based on existing input, almost all the criticisms of it, are criticisms that can apply to it's source material.
It's not really "intelligent" though, as in it's not thinking about what it's doing. What AI will do very well is reproduce jargon, and if it's jargon that we associate with intelligence then it appears intelligent. Academic papers for instance it can do a very convincing job because that format is so repetitive and jargon heavy.
You can do an experiment by asking it to produce a positive review of something niche and academic you're familiar with, then ask it to produce a negative review of the same subject. It will produce convincing dialogue for either scenario, but it does not know which is more true/accurate, and it will come across as a student writing about something they didn't do the reading for.
The "question if humans are [intelligent]" is the more relevant thing here. We're constantly expected to communicate with thoughtlessly reproduced jargon, and many of us can do this very well in a way that gives the impression of intelligent thought. The fact AI can do this, and that people are concerned about how intelligent it appears, is more a reflection on how derivative our notions of intelligence can be in these settings.
By its nature, Large Language Models won't ever be truly innovative, after all they rely on expected patterns. But a lot of the media that we consume is also made to appeal to patterns that we expect: genres, tropes, usual messages. AI could replace a lot of it and frankly, that's scary to think in a world where we need to work to earn our living.
Truly groundbreaking art may not be what people usually seek, it's often something they don't even know they want until they experience it, or they might even fail to appreciate it. But it likely won't be automated unless AI achieves full consciousness, but if it does we will have a much more complicated situation in our hands than "we can command AI to make art better than we can do ourselves".
Still, getting paranoid over the uncertain latter won't help us with the former that is just around the corner.
Good points.
One problem with replacing everything with AI that people don't think about: middle managers will start to be replaced too. There's no way to ask a LLM "why did you do that"? Fewer people will need to be managed.
It seems unwise to replace managers with LLMs because LLMs don't understand the real world implications of their responses, they don't have awareness of the real world, they simply give you often used language patterns, which can be innacurate or biased based on flawed human data. But it would be a great way for sketchy human executives to offload responsibility for unethical actions and feign objectivity or uninvolvement, so I don't doubt they will try.
Even if we imagine a perfect AI that does takes into account every objective fact and philosophical argument, that still leaves the question of how will the people who get replaced in all these intellectual, artistic and service jobs will make a living. That's not an answer that technology will give us, that will a nasty political situation.
No, you misunderstood. The managers are fired because there's fewer people to manage.
That makes sense too. Overall, a lot of people's jobs are threatened, but I don't think "learn AI" is going to cut it this time. Not for all these people.
Everyone in these threads likes to talk about being impressed by these llm or not being impressed by them as being some sort of intelligence test. I think of it more as a test of a person's sense of creativity.
It spits out a lot of passable text very easily, but as you're saying here its creativity is essentially nil. Even its "hallucinations" are just versions of things it borrowed from elsewhere injected slightly to wildly out of context in order to satisfy a prompt.
I tried to play a generative AI RPG builder game online and it came up with scenarios so boring I can't imagine playing it for longer than ten minutes.
I also find the same with generated content in other video games. At its best it's passable and that's about it. No man's sky has infinite worlds full of weird ligar creatures and after you've visited a couple dozen worlds they're pretty much all the same.
And who is to say that we humans don't process creativity exactly the same way? By borrowing from things we encounter.
Even the earliest creative expats of humans was just things we saw in nature, which we drew on cave walls.
We humans just have more experience since we existed longer, so the line feels a lot more blurred.
I also encountered games made by humans that were so boring I couldn't manage more than 10 minutes.
That's part of it, but it's definitely not all of it.
There's more creativity in the average prompt than there is in any response I've ever seen from ChatGPT.
If creativity were as simple as mashing a few things together as you're saying, ChatGPT would be there already because that's obviously what it's doing.
Me too, but that's an indictment of a single creator or team's idea that was boring, not an indictment of a system. This thing was basically a framework with the llm being the central "creator" at the center. It would find the most boring aspects of the prompts and lean into them. This is of course a subjective assessment, but I'd argue that it's not an uninformed one.
Minecraft would like to have a word with you...
Minecraft isn't generating new animals or narrative. Landscape generation is relatively straightforward from an algorithm / computation perspective. If it started generating its own models or characters or character dialogue I suspect it would very quickly fall into the territory of what I'm talking about.
There's just a feeling of emptiness to me that's pervasive in games with main parts of narrative or gameplay that are randomly generated.
I think the breakthroughs in AI have largely happened now as we're reaching a slowndown and an adoption phase
The research has been stagnating. Video with temporal consistency doesn't want to come, voice is still perceptibly non-human, openai is assembling 5 models in a trenchcoat to make gpt do images and it passing as progress, ...
Companies and people are adopting what is already there for new applications, it's getting more common to see neural network models in lots of solutions where the tech adds good value and is applicable, but the models aren't breaking new grounds like in 2021 anymore
The only new fundamental developments i can recall in the core technology is the push for smaller models trainable on way less data and that can be specialized for certain applications. Far away from the shock we all got when AI suddenly learned to draw a picture from a prompt
Wow, this is some spectacular hyperbole!
And ridiculously difficult to copyright any of it because it was generated.
Yes, AI doesn't work with copyright.
And since AI is here to stay, we better replace our failed copyright system with something proper. Disney be damned.
I'd like that? But if you're expecting the "we" in here to be the current people in their current power structures I suspect you'll be waiting an awfully long time for that result.
Here is an alternative Piped link(s):
results so far
robots that interact with the real world
Piped is a privacy-respecting open-source alternative frontend to YouTube.
I'm open-source; check me out at GitHub.
I want to note that everything you talk about is happening on the scales of months to single years. That's incredibly rapid pace, and also too short of a timeframe to determine true research trends.
Usually research is considered rapid if there is meaningful progression within a few years, and more realistically about a decade or so. I mean, take something like real time ray tracing, for comparison.
When I'm talking about the future of AI, I'm thinking like 10-20 years. We simply don't know enough about what is possible to say what will happen by then.