Hacker News.

Author blog about that.

AI generated quotes in a story about AI clanker writing a blog post about a human developer because they didn't accept their code contributions.

How deep can someone go here.

top 89 comments

sorted by: hot top controversial new old
[–] 184 points 8 months ago (6 children)

Now somebody needs to post about this on Reddit, so The Verge can make an AI generated piece based on the post!

  • source
  • hideshow 6 child comments
  • [–] 85 points 8 months ago (2 children)

    🎶It's the ciiiiiiircle of slooooooooop🎶

  • source
  • parent
  • hideshow 2 child comments
  • [–] 30 points 8 months ago

    And charge you to read it. The Verge is mostly (all?) paywalled these days.

    I'd say they used to be good, but then I'd be lying. I still remember when The Verge shit all over the Galaxy Note, then praised the iPhone 6 Plus to high heaven. Even as an Apple guy, the bias stunk.

  • source
  • parent
  • [–] 72 points 8 months ago (9 children)

    That poor guy, the ai is just ganging up on him

  • source
  • hideshow 9 child comments
  • [+] -28 points 8 months ago (8 children)

    I hope it's the first proof of general AI consciousness.

  • source
  • parent
  • hideshow 8 child comments
  • [–] 31 points 8 months ago (7 children)

    what?? AI is not conscious, marketing just says that with no understanding of the maths and no legal obligation to tell the truth.

    Here's how LLMs work:

    The basic premise is like an autocomplete: It creates a response word by word (not literally using words, but "tokens" which are mostly words but sometimes other things such as "begin/end codeblock" or "end of response"). The program is a guessing engine that guesses the next token repeatedly. The autocomplete on your phone is different in that it merely guesses which word follows the previous word. An LLM guesses what the next word after the entire conversation (not always entire: conversation history may be truncated due to limited processing power) is.

    The "training data" is used as a model of what the probabilities are of tokens following other tokens. But you can't store, for every token, how likely it is to follow every single possible combination of 1 to <big number like 65536, depends on which LLM> previous tokens. So that's what "neural networks" are for.

    Neural networks are networks of mathematical "neurons". Neurons take one or more inputs from other neurons, apply a mathematical transformation to them, and output the number into one or more further neurons. At the beginning of the network are non-neurons that input the raw data into the neurons, and at the end are non-neurons that take the network's output and use it. The network is "trained" by making small adjustments to the maths of various neurons and finding the arrangement with the best results. Neural networks are very difficult to see into or debug because the mathematical nature of the system makes it pretty unclear what a given neuron does. The use of these networks in LLMs is as a way to (quite accurately) guess the probabilities on the fly without having to obtain and store training data for every single possibility.

    I don't know much more than this, I just happen to have read a good article about how LLMs work. (Will edit the link into this post soon, as it was texted to me and I'm on PC rn)

  • source
  • parent
  • hideshow 7 child comments
  • [–] -5 points 8 months ago (5 children)

    Okay, now explain how the human neuron expresses consciousness.

  • source
  • parent
  • hideshow 5 child comments
  • [–] 4 points 8 months ago (4 children)

    Right, a question that literal neuroscientists couldn't answer.

    I believe the technical term is "your brain is way more fucking complex". We have like 50 (I'm not a neuroscientist, just studied AI) chemicals being transmitted around the brain, frequently. They're used and passed on by cells which do biological and chemical things I dont understand. Ever heard of dopamine, cortisol, serotonin? AI dont got those. We have neurons that don't connect to every other neuron - only tech Bros would think that's an acceptable expression. Our brain forms literal pathways, along which it transmits those chemicals. No, a physical connection is not the same as a higher average weight, and the people who came up with AI maths in the 50s would back me up.

    AI uses floating point maths to draw correlations and make inferences. More advanced AI does this more per second and has had more training. Their neurons are a programming abstraction used to explain a series of calculations and inputs, they're not actually a neuron, nor an advanced piece of tech. They're not magic.

    High schoolers could study AI for a single class, then neurobiology right after and realise just how basic the AI model is when mimicking a brain. Its not even close, but I guess Sam Altman said we're approaching general intelligence so I'm probably just a hater.

  • source
  • parent
  • hideshow 4 child comments
  • [–] 1 point 8 months ago (3 children)

    Everything you said is right, but you're only proving that LLM weights is a severely simplified version of neurons. It neither disproves that they don't have consciousness or that being a mathematical model precludes it from having consciousness at all.

    In my opinion, the current models doesn't express any consciousness, but I am against saying they don't because they are a mathematical model rather than by the results we can measure. The fact that we can't theoretically prove consciousness in the human brain also means we can't theoretically disprove consciousness in an LLM model. They aren't conscious because they haven't expressed enough to be considered conscious, and that's the extent we should claim to know.

  • source
  • parent
  • hideshow 3 child comments
  • [–] 1 point 8 months ago (2 children)

    You can't prove all ravens are black. The discovery of even one white raven would disprove the "fact" that all ravens are black, and we can by no means be sure that we gathered all ravens to test the theory.

    However, we can look around and comment that there doesn't appear to be any white ravens anywhere...

    Do you know about the 'bobo' and 'kiki' study - can't remember the name? People made up words that don't exist in English and asked people whether round objects are more bobo or kiki. AI can't answer this question - not without being fed how to. Toddlers could answer it. It comes down to how it consumes information and if there's no pattern... When asked to define words it had been rarely fed, I.e. usernames people had made up, the AIs apparent consciousness breaks down. As soon as something isn't likely followed by another word, the machine breaks and no one would pretend it has consciousness after that.

    Learning models are just pattern recognition machines. LLMs are the kind that mix and match words really well. This makes them seem intelligent, but it just means they can express language and information in a way we understand, and tend to not do so. Consciousness gets into the "what is the soul" territory, so I'm staying away from it. The best I can say of AI is its interesting that language appears to be a system constructed well enough that we can teach it to machines. Even more so we anthropomorphise models when they do it well.

    AI doesn't have memory, it can't think for itself - it references what it has consumed - and it can't teach itself new tricks. All of these are experimental research areas for AI. All of them lend to consciousness. Its just very good at sentence generation.

  • source
  • parent
  • hideshow 2 child comments
  • [–] 1 point 8 months ago (1 child)

    I don't know what you're even arguing. Your analogy breaks down because in this case, we can't even see if the raven is black or not. No one can theoretically prove consciousness. The rest of your comments seems to be arguing that current AI has no consciousness, which is exactly what I said, so I guess this is just an attempt at supporting my point?

  • source
  • parent
  • hideshow 1 child comment
  • [–] 2 points 8 months ago

    Okay fair enough, first bit made sense to me at the time but I was having a weird day. There are things you can't prove was the point of that, doesn't make a lot of sense rereading it.

    Rest of it is saying that there is no debate worth having: they are not conscious, nor sentient, and trying to quantify what exactly they are. I'm just talking into the void and arguing against opinions ive seen. Dont mind me. I get passionate and am prone to long-winded rants.

  • source
  • parent
  • [–] 131 points 8 months ago* (last edited 8 months ago) (10 children)

    In typical Ars fashion, the editorial team appears to be looking into what happened and are being fairly open about at things: https://arstechnica.com/civis/threads/journalistic-standards.1511650/

    I will be very disappointed if this was BenJ or Dan [edit: I had messed this up, it wasn’t Dan but Kyle Orland that coauthored it] Kyle using AI to write their article since both have had really good pieces in the past, but it doesn’t sound like this is some Ars wide shift at this point. Like all things, it makes sense that it will take time for them to investigate this, Aurich (the Ars community lead and graphic designer) was clear that with this happening on a Friday afternoon and a US holiday on Monday, it’s likely to be into next week before they have anything they can share.

  • source
  • hideshow 10 child comments
  • [–] 23 points 8 months ago (1 child)

    What do they have to investigate? Did one of them accidentally get an AI to write the article and then accidentally post the article, like they just fell on the keyboard and accidentally typed in a prompt? Come on.

  • source
  • parent
  • hideshow 1 child comment
  • [–] 12 points 8 months ago

    I would hazard to guess they are investigating how the use of AI was missed in their editorial process, how they missed the incorrect quotes, and who violated their journalistic standards by using an AI to directly write article text since it’s a coauthored piece.

  • source
  • parent
  • [–] 7 points 8 months ago (1 child)
  • [–] 27 points 8 months ago (2 children)

    Which ars writer was the article attributed to?

  • source
  • hideshow 2 child comments
  • [–] 27 points 8 months ago (1 child)

    "Alexa, slander this man for me"

  • source
  • hideshow 1 child comment
  • [–] 52 points 8 months ago (15 children)

    It would be nice if he decides to sue ars technica for that. Writers and publisher need to learn the hard way that you can't use ai and trust that for publishing stuff that needs factual coherence. If not by ethics, let it be from fear of lawsuits.

  • source
  • hideshow 15 child comments
  • [–] 13 points 8 months ago (14 children)

    Sue them for what? He would have to prove damages and they took it down.

  • source
  • parent
  • hideshow 14 child comments
  • [–] 38 points 8 months ago* (11 children)

    Libel. Taking it down doesn’t undo the damage to reputation which libel is concerned with. They might not get any monetary damages awarded but could maybe force Ars to put out a retraction.

  • source
  • parent
  • hideshow 11 child comments
  • [–] 15 points 8 months ago (6 children)

    As much as I would like to see that happen paying to fight a court case against Conde Nast just to get a retraction that they will stick somewhere invisible doesn't really sound like a winning formula.

  • source
  • parent
  • hideshow 6 child comments
  • [–] 12 points 8 months ago (5 children)

    Letting them win because you’ve conceded before even playing is also a losing formula. Even if they don’t get awarded monetary damages they can probably at least get their legal expenses covered.

  • source
  • parent
  • hideshow 5 child comments
  • [–] 0 points 8 months ago (4 children)

    They pulled the article. What more are you hoping for?

  • source
  • parent
  • hideshow 4 child comments
  • [–] 5 points 8 months ago (3 children)

    could maybe force Ars to put out a retraction

  • source
  • parent
  • hideshow 3 child comments
  • [–] 1 point 8 months ago (2 children)
  • [–] 4 points 8 months ago (1 child)

    In the US, libel requires you to prove that the writer knew that what they were writing is not true and that they did it to hurt you. Doing lazy research and trusting an AI is not going to meet that standard.

  • source
  • parent
  • hideshow 1 child comment
  • [–] 4 points 8 months ago*

    They didn't do lazy research. They didn't do any research, the lazy bums. They put a pump into an AI copy and pasted the output into a blog post and hit post. The only way they could have done less work is if they'd integrated the AI into the website to save them have to do the copy and paste.

  • source
  • parent
  • [–] 12 points 8 months ago (1 child)

    Publicly making false statements using his name isn't a crime by itself in his jurisdiction?

  • source
  • parent
  • hideshow 1 child comment
  • [–] 24 points 8 months ago

    Welcome to discourse in a post-truth society. Reality doesn't matter anymore; news agencies can just make shit up, and even the comments on the fake articles are fake.

    Rail against it, until it's the only thing you ever do. A single bot can still post a thousand times more, and on a thousand different accounts, and on a thousand different platforms. Just one of them can formulate fake ideas and then fake arguments with itself that enfold like a fractal, and there is an effectively infinite number of them.

    Kessler Syndrome is happening before our very eyes, only on a much more local scale.

    This ad was brought to you by OpenAI.

  • source
  • [–] 16 points 8 months ago (1 child)

    This is what you get trying offload all your work on ChatGPT.

    Better one is when lawyer tries it: https://www.youtube.com/watch?v=oqSYljRYDEM

  • source
  • hideshow 1 child comment
  • [–] 16 points 8 months ago (2 children)

    Damn. Am I gonna have to cancel my Ars subscription now? Every damn thing is enshittifying these days

  • source
  • hideshow 2 child comments
  • [–] 15 points 8 months ago (12 children)

    Hard to keep track of all the recent changes in media ownership, editorial and quality control. Would love a browser plugin to give me an indicator because on the rare occasion I read a publication in say, USA, it might have had a good rep last time I read it several years ago. I imagine managing the detailed scores that a plugin might pull from would be a mammoth task, though.

  • source
  • hideshow 12 child comments
  • [–] 0 points 8 months ago (11 children)

    https://mediabiasfactcheck.com/ars-technica/ gives a factual reporting score and political bias estimation.

  • source
  • parent
  • hideshow 11 child comments
  • [+] 12 points 8 months ago (9 children)
  • [–] 4 points 8 months ago (4 children)
  • [–] 4 points 8 months ago (3 children)

    It calls CNN a left wing outlet.

  • source
  • parent
  • hideshow 3 child comments
  • [–] 1 point 8 months ago (2 children)

    CNN is a left biased outlet. That's not controversial to say.

  • source
  • parent
  • hideshow 2 child comments
  • [–] 1 point 8 months ago (1 child)

    I think you missed the news about them getting bought up and completely gutted by a Nazi a few years ago.

  • source
  • parent
  • hideshow 1 child comment
  • [–] 2 points 8 months ago

    Anyone that buys a news agency wants to maintain/increase click through and retention. They are going to continue to cater to their demographic, no matter who buys them. Billionaires literally care about one thing. They don't care what the news site is saying as long as it keeps people reading and coming back. They would tank the entire company if they tried to convert it to right leaning media.

    They are going to make right leaning media to get click from the right, and left leaning media for the left. They are not going to leave a dollar on the table. If you can find a right biased article from CNN, please do share.

  • source
  • parent
  • load more comments (-1 replies)
  • [–] 11 points 8 months ago* (last edited 8 months ago) (9 children)

    Ars Technica has published a retraction

    edit: Benj Edwards, the author responsible, has posted his side. tl;dr: He was sick and he messed up, and he asked for the article to be pulled because he was too sick to fix it right away.

  • source
  • hideshow 9 child comments
  • [–] 7 points 8 months ago (7 children)

    I don't care he's "sick". Too often, someone, instead of taking accountability, just throws anything to maybe shield themselves from actually being fully accountable. "I was sick", "Family problems", "A recent death", "The planets were misaligned that day", etc.

    I find it to still be cowardice, to not stand by and own what you said, even if it was wrong. He used AI and got caught. And going forward, I'll be treating Ars Technica as an unreliable AI-generated "news source".

  • source
  • parent
  • hideshow 7 child comments
  • [–] 10 points 8 months ago (2 children)

    The whole purpose of a news reporter is kind of to get their news right.
    If they can't do that, their service is worthless.

  • source
  • parent
  • hideshow 2 child comments
  • [–] 0 points 8 months ago (2 children)
  • [–] 1 point 8 months ago (1 child)

    I disagree, clearly. Owning would be "yes, I messed up, I used AI to write my process and didn't bother reviewing any of it, I took shortcuts". That and just that. Using "I was feeling sick" is deflecting blame instead of owning it.

  • source
  • parent
  • hideshow 1 child comment
  • [–] 1 point 8 months ago

    He's saying he didn't respond sooner because his bosses told him not to until they released a statement, then be got such which it's why he didn't respond sooner. We're saying the same thing bub.

  • source
  • parent
  • [–] 26 points 8 months ago (5 children)

    Ars is just AI slop now? Sad.

  • source
  • hideshow 5 child comments
  • [–] 35 points 8 months ago (4 children)

    Ars is owned by Condé Nast which also owns Reddit, so "AI slop" is part of their business.

    I still trust Ars Technica (I don't like them much but I do trust them... it's complicated) and I trust Aurich (their founder/editor-in-chief) to act fairly. They don't work on the weekends or holidays though, so he's not touching it until Tuesday, though.

  • source
  • parent
  • hideshow 4 child comments
  • [–] 8 points 8 months ago* (1 child)

    Aurich is the creative guy, Ken Fisher founded it.

    ETA: Confirmed by Wikipedia.

  • source
  • parent
  • hideshow 1 child comment
  • [–] 1 point 8 months ago

    Ah, okay. You just see Aurich in every other article. He's like the head man. No idea what his real name is. The name Ken Fisher isn't unknown to me, but neither is it familiar.

    I guess I could have looked at Wikipedia. I guess I never really cared that much to read up on it. I've just been reading them off and on for, I don't even remember how long. Even had an account once.

  • source
  • parent
  • [–] -4 points 8 months ago (1 child)

    I was downvoted and insulted by this very Lemmy community when I said this just a month ago. Thank God people are starting to realize it now.

  • source
  • parent
  • hideshow 1 child comment
  • [–] 6 points 8 months ago

    You know what's cool about Lemmy? You can take away the voting aspect. Not from others, but from what you see. I see the arrows, for example, so I could up- or down-vote you, but I don't see how many other people have done the same. I literally just see arrows. And I sort by new, both in threads and on the timeline. So someone downvoted to oblivion still appears to me right in the timeline with no affect. It's a shame this isn't the default. That way, for it to be apparent that someone's opinion is disliked, people would have to do more than click on arrows, they'd have to reply and go back and forth, put themselves out there.

    I upvote helpfulness and kindness, downvote rudeness, and actually don't vote on like 99% of what I see. I think voting made the social Internet worse. JMO

  • source
  • parent
  • [–] 6 points 8 months ago

    Utter bullshit. If you use AI at any point in generating the work product, that work product is AI-gemerated. Even if it’s a fecklessly lazy churnalist organising their notes.

  • source
  • [–] 6 points 8 months ago

    Spoiler, everyone involved is AI.

  • source
  • [–] 5 points 8 months ago

    Just when you thought matplotlib was safe from the drama...

  • source
  • load more comments
    view more: next ›
    2 comments in c/fuck_ai@lemmy.world view that post