1. 42
  1.  

  2. 23

    I read the first part and was going to post something snarky about bitcoin enthusiasts failing the Turing test…then I failed it myself. Well shit.

    1. 13

      It’s a little worrisome to me that people don’t see through this. It’s grammatically fine, but there is no conceptual power behind the words, which is a common problem (in different forms) with procedural generation algorithms. I would have hoped that readers expect, and can evaluate, coherence of writing beyond grammar. Maybe that’s overly optimistic.

      1. 26

        There are plenty of low-quality blog posts with no conceptual power behind their words.

        1. 4

          Can you imagine what the content marketing mills will do with this?

          1. 2

            Please don’t give them ideas; If I get a new story from Maupassant in my email, there is no way I am not going to read it.

            1. 1

              They’re just replacing humans with robots, which has been the trend for decades in other industries.

            2. 3

              I saw a tweet about jokes generated by GPT-3 being amazing, complete with a screenshot. When I clicked through I found Gwern’s tests. Nothing like what was implied. None of the examples in the screenshot were actual output.

              It was the No Man’s Sky of joke writing. Semantically correct but an empty void when it came to humour.

              1. 2

                Yeah, I got about halfway through the generated post and then just started scrolling, which led me straight to the gotcha. I was completely bored and confused by what the ‘author’ was trying to say. Sure, it was largely grammatically coherent, but bits of it just flat out did not make sense in context. I was especially confused by the line ‘The forum also has many people I don’t like’, which I read about three times trying to work out why the ‘author’ had said that. There were several other bits that tripped me up because they just didn’t make contextual sense. While I could see this tool being useful for generating painting-by-numbers style sports reports and the like I’m not too worried about it putting actual writers with something to say out of a job.

                1. 1

                  ‘The forum also has many people I don’t like’, which I read about three times trying to work out why the ‘author’ had said that.

                  That was my favorite part. I read it as the bot saying “There’s a lot of people I don’t like, and I’ll show everyone how dumb they are when they think a bot is being intelligent and funny.”

                2. 1

                  It’s grammatically fine, but there is no conceptual power behind the words, which is a common problem (in different forms) with procedural generation algorithms.

                  And it’s a common problem in people too. Too many lost points on high-school essays…

                3. 5

                  Hah, this reminds me of an XKCD comic: https://xkcd.com/810/

                  Is this a problem if the content that the AI generates is a worthwhile read?

                  It would be the biggest plot twist if the entire blog post was generated by GPT-3, including the last section.

                  1. 3

                    It’s a problem because the experiment didn’t happen. And unless someone fact checks all of its output, AI generated articles are going to lie a lot.

                  2. 4

                    Read the summary, metagamed, was right. Metagaming ftw!

                    The forum also has many people I don’t like. I expect them to be disproportionately excited by the possibility of having a new poster that appears to be intelligent and relevant.

                    GPT-3 is savage.

                    1. 4

                      After being fooled, believing that the words were written by a human, I took for granted that OpenAi’s GPT-3 was “released into the wild of open source”.

                      However, (GPT-3 was not expecting that :-)) when I go to the OpenAI FAQ page it says: “Why did OpenAI decide to release a commercial product?”

                      Should I believe OpenAI GPT-3 or OpenAI FAQ ?

                      It seems there is a contradiction between the impressive GPT-3 output, and reality.

                      1. 4

                        If you skim this quickly, as people often do, you try to pick out the interesting things and basically create your own narrative out of someone else’s.

                        If you read it slowly, you notice how unnatural this text is. It’s very repetitive in style. There are jumps in logic. There is a general lack of coherence.

                        Some will now respond “many people are also like that”. But all of GPT-3’s output is like that and no blogs I’m likely to bookmark or appreciate are like that.

                        If this is the future of tech support and news, it’s a dystopia to me.

                        1. 1

                          Placing a caveat here for possible future reference: I do believe that what the GPT models do can be useful and It definitely seems plausible that this capability will be an important part of a future AGI. I’m just appalled by the enthusiasm for the examples of its products that people are currently posting, as I find those unacceptable as replacements for human products.

                        2. 5

                          I explain why I think GPT-3 has disruptive potential comparable to that of blockchain technology.

                          So zero disruptive potential?

                          1. 2

                            Great potential to disrupt the investment strategies of some gullible and naïve technology enthusiasts! Not such a bad parallel, after all.

                          2. 2

                            I’m incredibly conflicted about whether to flag this or not.

                            … No, it should stay.

                            1. 2

                              The dimensionality of the written word is low enough that these methods really shine. It isn’t high art, and has a number of tells, but it did fool quite a few people!

                              The scammers are probably burning the midnight oil as we speak–to incorporate this non-profit wizardry into their spam farms. Nigerian princes are about to get an upgrade.

                              1. 1

                                I feel like I’ve been had :) I’d like to see the 9 earlier “drafts” as well.

                                1. 1

                                  Stuff like this is super impressive, but it also makes me wonder why software developers are so intent on putting themselves out of jobs.

                                  1. 5

                                    It’s the healthcare paradox: the majority of passionate doctors want to exist in a world where they don’t have jobs anymore because everybody’s healthy.

                                  2. -3

                                    the biggest thing since bitcoin

                                    Meaning, “fails at its primary (only) purpose and only useful for running Ponzi schemes, while accelerating climate change?”

                                    I haven’t read the article; the headline turned me off already.

                                    1. 4

                                      You might want to actually give it a shot. Take it from someone who hates these headlines too.

                                      1. 0

                                        I’ve read some (what I think will be) more nuanced posts on GPT-3. I guess it’s interesting, but I’m not really invested enough to have formed an opinion on this one (yet).

                                        1. 8

                                          No seriously, read the article.

                                          1. 11

                                            This article is a great litmus test for people who ignore or flag things based on the headline.

                                      2. 3

                                        How does Bitcoin fail at being a peer-to-peer electronic cash system?

                                        1. 1

                                          It’s too slow to replace cash. People do the bulk of the transactions off-chain, which kinda defeats the purpose.

                                          1. 3

                                            Your main criticism is “it’s too slow”? All digital money is slow. It only looks fast because banks take on the risk of digital money transfers and give you the benefit of the doubt. For “digital cash”, I’d say 10 minutes is pretty good.

                                            1. 4

                                              banks take on the risk of digital money transfers and give you the benefit of the doubt

                                              That’s kind of a killer feature, though.

                                              1. 3

                                                If you desperately need that kind of thing, yes. Bitcoin provides benefits traditional money and banking doesn’t, hence it’s existence. There is nothing preventing banking solutions on top of Bitcoin.

                                                1. 2

                                                  The primary benefits of Bitcoon are lack of regulation and high volatility due to same, and a secondary benefit of being distributed with no bias towards societal economic utility for the people getting lucky while mining.

                                              2. 1

                                                You just got done comparing it to cash, not debit or credit card transactions. Cash is instantaneous. Credit cards have fraud detection, which Bitcoin lacks.

                                                1. 2

                                                  How is cash instantaneous acorss the ocean?

                                                  1. 1

                                                    I’m not sure why I need to say this but transporting money is not the same as exchanging it

                                                2. 1

                                                  Most bank transfers days 2 days anyway