Hacker News Reader

The view from outside of tech is very much "We went from having no AI to having ChatGPT overnight! Imagine how soon the future will be here, we could have a breakthrough for AI at breakfast tomorrow and talking companion robots the day after that!"

And the doomers just add on "and the week after that we'll all be dead!"

I'm personally not very impressed by the AI tools I've used. Sure they're a neat toy. They do seem to keep getting better. Maybe they'll be good enough one day for me to actually want to use them and feel it's a benefit.

> The view from outside of tech is very much "We went from having no AI to having ChatGPT overnight!"

In their defense, for most people that is what happened. Sure, they've been using "AI" tools of varying degree for a long time (spellcheck, language translation), but now they have totally free access to something that behaves sorta like the AIs they've seen on the popular Star Trek shows -- where, incidentally, the AIs were also imperfect.

Journalists love their overnight success stories. And it's most apparent when they are talking about some musician who worked their ass off for decades before being discovered. As if they sprung into existence as 30 year-olds, worked hard for a couple years, then became millionaires.

> I'm personally not very impressed by the AI tools I've used. Sure they're a neat toy. They do seem to keep getting better. Maybe they'll be good enough one day for me to actually want to use them and feel it's a benefit.

Unless you explain this statement, most people here are likely to dismiss everything you have to say on the topic of AI.

We all know exactly the kinds of bad outputs we've seen from AI.

I just ran a query for ChatGPT to recommend various databases for a specific use case. Out of the seven databases it recommended, only one was a database actually appropriate; one suggestion was marginally acceptable; three of the recommendations weren't even databases.

I then asked it to provide a list of the most important battles in North Africa prior to the entry of the United States into World War 2.

It gave me five answers. Three of which occurred after the entry of the United States into World War 2.

AIs provides extremely plausible answers. Sometimes it will actually generate correct, useful output; but you cannot yet rely on it for correctness.

I'd like to see a side by side comparison with a random human on the street. Maybe with a sample size of 100 or so. How well do you think the humans would do vs whatever outdated model you were playing with here?

There is clearly significant value to this tech and I'm still dumbfounded how strongly some people try to deny it.

Anyone reckon there's a chance that GPT hallucinates because it was trained on online material (e.g. reddit and other forums)? I'd have to say on topics I know GPT is about as hit or miss as a random internet comment, especially in that they'll both give a confidently stated answer whether the answer is factual or not.

Is it possible GPT just thinks[0] that any answer stated confidently is preferable over not giving an answer?

Promise I'm not just being snarky, legitimate wonder!

[0]: I know it doesn't actually think, you know what I mean

You're judging fish by its ability to climb a tree. Being able to recall facts is a nice side effect for LLMs, not their bread and butter. If you need facts, plug in some RAG to it.

Also, what model did you use?

Stop using GPT 3.5 and complaining it's no good. We all know that. Unless you're using GPT4, your anecdotes are out of date and irrelevant.

https://chat.openai.com/share/43ebf64e-34ae-402b-a3ce-0787e2...

It is really why I am tired of this whole format of talking to people.

I wish someone would build a site that we share chatGPT4 outputs like the above.

I can't remember the last time I have actually learned something from a human on a message board like this compared to chatGPT4.

Talking to people like this is just 100% waste of time.

Well, learning and conversing are different things done for different reasons.

I'll agree that GPT4 has completely replaced Google, stackoverflow, etc for me.

The only time I use Google now is for janky more human like situations.

For example, today I had to transfer DC roles from a Windows 2012 R2 server to a new 2022. They only have one DC. And the old DC has a basically unused CA service set up.

ChatGPT would have me "fix" everything first, wheras, I did find a forum post with a situation almost identical to mine that helped me cowboy it rather than being overly meticulous.

There is still value to human experience. For now.

Unless you explain this statement, most people here are likely to dismiss everything you have to say on the topic of AI.

fwiw, I agree with them. It has its use cases, but 'hallucinations' or whatever you want to call them is a huge dealbreaker for everything I'd want to use ai for

Agreed but in my opinion the problem is more fundamental than just hallucinations, it involves plain inaccuracy and inability to reason.

Try asking chatgpt or Gemini about something complex that you know all about. You’ll likely notice some inaccuracies, or thinking one related subject is more important than something else. That’s not even scratching the surface of the weird things they do in the name of “safety” like refusing to do work, paying lip service to heterodox opinions, or interjecting hidden race/gender prompts to submodels.

It’s good at generalist information retrieval to a certain degree. But it’s basically like an overconfident college sophomore majoring in all subjects. Progressing past that point requires a completely different underlying approach to AI because you can’t just model text anymore to reason about new and unknown subjects. It’s not something we can tweak and iterate into in the near term.

This same story has recurred after every single ML advance from DL, to CNN + RNN/LSTM, to transformers.

> Agreed but in my opinion the problem is more fundamental than just hallucinations, it involves plain inaccuracy and inability to reason.

> Try asking chatgpt or Gemini about something complex that you know all about. You’ll likely notice some inaccuracies, or thinking one related subject is more important than something else. That’s not even scratching the surface of the weird things they do in the name of “safety” like refusing to do work, paying lip service to heterodox opinions, or interjecting hidden race/gender prompts to submodels.

For sure.

On the other hand, I recently started to convert these AI hallucinations into a feature for me: it is like asking a person who is somewhat smart, but high on some hallucinogenic drug, on their opinion on a topic of your interest. Depending on the topic and your own intellectual openness, the result can be ... interesting and inspiring.

I’ve asked Bing about sexual health questions and been told it’s not okay to talk about that.

I asked it a question about Christianity and it stated things with the tone and certainty of a preacher.

Just gross. And worse to put something like that in the hands of billionaires.

I am somewhat of a doomer not because I think it will be the Terminator. More like AI theocracy here we go.

Generally agree with the GP, and am curious what use-cases you've found where AI meaningfully improves your daily work.

I've found two so far: the review summaries on Google Play are generally quite accurate, and much easier than scrolling through dozens of reviews, and the automatic meeting notes from Google Meet are great and mean that I don't have to take notes at a meeting anymore.

It did okay at finding and tabulating a list of local government websites, but had enough of an error rate (~10%) that I would've had to go through the whole list to verify its factualness, which defeats a lot of the time savings of using ChatGPT.

Beyond that: I tried ChatGPT vs. Google Search when I had what turned out to appendicitis, asking about symptoms, and eventually the 5th or so Google result convinced me to go in. If I had followed ChatGPT's "diagnosis", I would be dead. I've tried to have ChatGPT write code for me; it works for toy examples, but anything halfway complicated won't compile half the time, and it's very far from having maintainable structure or optimal performance. Basically works well if your idea of coding is copying StackOverflow posts, but that was never how I coded. I tried getting ChatGPT to write some newspaper articles for me; it created cogent text that didn't say anything. I did some better prompting, telling to incorporate some specific factual data - it did this well, but looking up the factual data is most of the task in the first place, and its accuracy wasn't high enough to automate this task with confidence.

Bard was utter crap at math. ChatGPT is better, but Wolfram Alpha or just a Google Search is better still.

In general, I've found LLMs to be very effective at spewing out crap. To be fair, most of the economy and public discourse involves spewing out crap these days, so to that extent it can automate a lot of people's jobs. But I've already found myself just withdrawing from public discourse as a result - I invest my time in my family and local community, and let the ad bots duke it out (while collecting a fat salary from one of the major beneficiaries of the ad fraud economy).

I recognize your username so I know you've been around for awhile (and are you a xoogler who for a time banged the drum on the benefits of iframes, or am I confusing you with a similar username?), and so I'm kind of surprised at your lukewarm take on LLMs.

I agree they hallucinate and write bad code and whatever, but the fact that they work at all is just magical to me. GPT-4 is just an incredibly good, infinitely flexible, natural language interface. I feel like it's so good people don't even realize what it's doing. Like, it never makes a grammatical mistake! You can have totally natural conversations with it. It doesn't use hardcoded algorithms or English grammar references, it just speaks at a native level.

I don't think it needs to be concretely useful yet to be incredible. For anyone who's used Eliza, or talked to NPCs, or programmed a spellchecker or grammar checker, I think it should be obviously incredible already.

I'm not sold on it being a queryable knowledge store of all human information yet, but certainly it's laying the inevitable future of interacting with technology through natural language, as a translation layer.

> GPT-4 is just an incredibly good, infinitely flexible, natural language interface.

An interface that's incredibly difficult to produce consistent output. As far as I know, we have not found a way to even make it do basic tasks parsed from natural language without a prohibitive-to-most-use-cases error rate. It's amazing that it can produce pretty believable looking text, but it's abundantly clear that there's no reasoning behind that text at all.

The other day I planned out a cloud to on-prem migration of an entire environment. From cost analysis to step by step checklists. In about 2 hours I had a ~50 page run book that would have taken me at least a week to do coming from my own brain and fingertips.

How? Are you letting it make decisions for you?

Here is my initial draft chat session. From here I feed it parts of this initial thing. It gets something down on the page immediately and I revise myself and by feeding portions into new chat sessions etc.

https://chat.openai.com/share/15b30c88-d21f-4ffe-8c15-5b444d...

Good reminder than no social media platform is a monolith. Trying to speak as the voice of a platform typically gets you egg on your face, especially when being dismissive towards someone else.

You’ll find people who claim to have doubled their productivity from ChatGPT and people who think it’s useless here.

That’s their loss isn’t it?

Count me unimpressed too

Bing Copilot can’t even answer simple questions about businesses in a particular city without lying. When confronted, it will apologize and then repeat the same text verbatim.

Zero reasoning happening.

> the fundamentally unsustainable nature of industrial society

Nothing that goes against entropy is fundamentally sustainable. That doesn’t mean we can’t keep it going for time that is essentially infinite on human scales (unsustainable life has been kicking around on Earth for 3.7 billion years now). Defeatism is even more dangerous than hallucinating optimism.

Do you mind explaining what your comment means in layman's terms?

Every process of life - breathing, moving, learning - takes energy. This is as true for the ants building their colonies as it is for us building our cities.

Due to thermodynamics, this is a one-way process. In our case, it’s all fueled by the sun, which will keep burning for another few billion years.

But even that counts being not not truly sustainable, since one day it must end. The whole universe is destined to die a cold heat death where nothing at all happens anymore.

But that is a long time away and until then, we can build a beautiful civilization; we can learn and grow; and we can do so with nuclear & solar power.

People saying that we’re all as good as dead are technically correct, but in a very unhelpful way.

> Tech can't stand "bad thoughts". > it's like they've all taken obscene quantities of some new stimulant and can't even connect with reality anymore

It's hallucinogens. That's the drug of choice in tech. What that leads to, other than the obvious hallucinations, is an obsession with set and setting. Bad vibes will literally give you a bad trip and ruin your high. I've seen this with wealthy shroom heads over and over again. Their entire life becomes centered around making sure the set and setting are always perfect, which means any little bit of someone trying to talk some sense in to them, they get ignored.

Once you start to think through the kinds of behaviors that shroom addicts would start to engage in, especially if they had the wealth and resources to facilitate their addiction, you'll see it everywhere. It's not the typical "ruin your life" kind of addiction, but it's having an impact on what kind of ideas are allowed among the SV venture types.

I couldn't speak for silicon valley, but the personality changes you're describing as being associated with shrooms... those are not typical.

I have a few friends in psychedelic assisted therapy and the effects I've noticed in them are the same effects I've noticed with regular psychedelic use outside of therapy: you don't identify the bad vibes and shy away from them, you end up making decisions that are uncomfortable in the sort term to improve things in the long term. Myself, I started going to college.

Maybe it works differently among wealthy people.

If you're in a cult of positivity, adding psychedelics to the mix is more likely to make you acutely away of the inauthenticity of the situation.

Psychedelic assisted therapy is so good an powerful precisely because it exploits the best case scenario of set and setting with the way the drug has an impact on your mind. I'm far from against psychedelic and have done the therapy myself.

However, if you're in a positivity cult, and don't realize you're in a cult, and don't have someone guiding you to consider you might be in a cult, the shrooms are just as likely to make the cult seem like the most profound and important experience in your entire life.

If you are a lead in a company, you suddenly have a profound spiritual experience based around your ability to hire and fire people and tell them what to do and can use the drug to convince yourself that the ideas you are coming up with are the most profound thoughts a person has ever had. You won't even realize you're shutting out good ideas, because you have a messianic belief in AI or crypto or whatever the thing is, and you take the shrooms to reinforce that belief and you create an environment around yourself and put people around you that reinforce that belief.

It's a very different experience than going to therapy to work through your fear or depression.

> It's hallucinogens. That's the drug of choice in tech

To the degree Silicon Valley has a drug right now, it’s ketamine. (Before that, it was weed.)

HGH and low dosage testosterone regimes seem popular as well.

Seriously? That just seems ridiculous. Is this an aging thing, where folks are worried about low testosterone or something?

A decent amount of AI doomerism I've seen is classic criti-hype (as coined in [1], afaik)--it's promotion of the technology thinly disguised as critique. Fear mongering about AIs taking artists' or writers' job is yet another way of boasting about their capabilities.

[1] https://sts-news.medium.com/youre-doing-it-wrong-notes-on-cr...

"My product is so good the world should be afraid of it" often comes from the mouths of the AI CEOs.

Same with CEOs who say "Our tech is too powerful, we want to be regulated" while knowing that Congress can barely function to pass a budget.

I was really starting to feel like Abe Simpson at one point and then I read From Good to Great. The central thesis of that book is that you might work at a good company that avoids confronting 'brutal facts' but you will never find a Great company that does.

You have to slay bad things to make a great company, and you have to identify them before they're chewing on something vital. Which means going out of your way to speculate about whether things are actually bad or just annoying.

> when I talk to people outside of my area it's like they've all taken obscene quantities of some new stimulant and can't even connect with reality anymore

Can you share some examples of what you mean by this? I've encountered people who are excited about "AI", for sure, but who are excited because a problem that has plagued them for years suddenly became solvable. Excited because the way they used to learn about things just suddenly changed. Excited because there's a better way to do research than to scour the crappy results provided by the big search engines.

There is absolutely an element of hype that far outstrips the reality, and after the single-function apps like "Summarize this paper for me" are monetized to death, we'll enter the trough of disillusionment, and some reality will set in. But there are absolutely transformative use cases unlocked by the latest generation of tools that are very real, exciting, and that enable a new generation of users to interact with computers in a way that was previously the stuff of science fiction.

I've spent much of my career working on boring enterprise tech, building software that's not flashy, but gets work done. The problem space is extremely large, and the solution space is extremely inadequate. It's this vast space of unsolved business problems that make some of this hype more real, IMO. LLMs in particular will be transformative for many of the problems the big enterprise platforms solve, and have the potential to solve some of the messiest parts of operating in that space.

I don't know that I agree with the "Years of gloom" characterization, and I agree that there are other underlying currents re: sustainability, but I can't help but feel this comment endorses a different problematic extreme. In the middle of the hysteria are real uses cases that will change how we interface with computers, and industries will be transformed/new categories created.

Anecdotally, GPT4 has completely changed how I approach research and troubleshooting and has been saving me many hours on a regular basis. Weird error message in a linux log file? I'll provide some context and paste the error, and a few minutes later the issue is solved based on a decently good answer.

Picking up a new language? I'll ask for example code to solve a very specific use case, and then keep returning when I run into errors. The rate of learning enabled this way is pretty remarkable.

If the only thing that we get out of this craze are some extremely high quality LLMs with an accurate understanding of most human knowledge, that alone is an incredible jump forward.

Exactly. The ai car ethics news was always marketing to distract from the fact that it doesn’t work.

As someone who works "close to the metal" in AI, do you really believe AI is just another "bubble" the way crypto was? I'm having a tough time finding the two even similar. It's clear how much potential it has to change things far more than even the most optimistic views of what crypto bros were touting as the future a few years ago.