Just as the internet was a democratization of information, llms are a democratization of output.
That may be in terms of production or art. There is clearly a lower barrier for achieving both now compared to pre-llm. If you can't see this then you don't just have your head stuck in the sand, you have it severed and blasted into another reality.
The reason why you reacted in such a way is again, a lack of imagination. To you, "work" means "employment" and a means to a paycheck. But work is more than that. It is the output that matters, and whether that output benefits you or your employer is up to you. You now have more leverage than ever for making it benefit you because you're not paying that much time/money to ask an LLM to do it for you.
Pre-llm, most for-hire work was only accessible to companies with a much bigger bank account than yours.
There is an ungodly amount of white collar workers maintaining spreadsheets and doing bullshit jobs that LLMs can do just fine. And that's not to say all of those jobs have completely useless output, it's just that the amount of bodies it takes to produce that output is unreasonable.
We are just getting started getting rid of them. But the best part of it is that you can do all of those bullshit jobs with an LLM for whatever idea you have in your pocket.
For example, I don't need an army of junior engineers to write all my boilerplate for me. I might have a protege if I am looking to actually mentor someone and hire them for that reason, but I can easily also just use LLMs to make boilerplate and write unit tests for me at the same time. Previously I would have had to have 1 million dollars sitting around to fund the amount of output that I am able to produce with a $20 subscription to an LLM service.
The junior engineer can also do this too, albeit in most cases less effectively.
That's democratization of work.
In your "5% unemployment" world you have many more gatekeepers and financial barriers.
I write code to drive hardware, in an unusual programming style. The company pays for Augment (which is now based on o4, which is supposed to be really good?!?). It's great at me typing: print_debug( at which point it often guesses right as to which local variables or parameters I want to debug - but not always. And it can often get the loop iteration part correct if I need to, for example, loop through a vector. The couple of times I asked it to write a unit test? Sure, it got a the basic function call / lambda setup correct, but the test itself was useless. And a bunch of times, it brings back code I was experimenting with 3 months ago and never kept / committed, just because I'm at the same spot in the same file..
I do believe that some people are having reasonable outcomes, but it's not "out of the box" - and it's faster for me to write the code I need to write than to try 25 different prompt variations.
Thanks for sharing your perspective with ACTUAL details unlike most people that have gotten bad results.
Sadly hardware programming is probably going to lag or never be figured out because there's just not enough info to train on. This might change in the future when/if reasoning models get better but there's no guarantee of that.
> which is now based on o4
based on o4 or is o4, those are two different things. augment says this: https://support.augmentcode.com/articles/5949245054-what-mod...
Augment uses many models, including ones that we train ourselves. Each interaction you have with Augment will touch multiple models. Our perspective is that the choice of models is an implementation detail, and the user does not need to stay current with the latest developments in the world of AI models to fully take advantage of our platform.
You might be getting bad results for a few reasons:
- your prompts are not specific enough
- your context is poisoned. how strategically are you providing context to the prompt? a good trick is to give the llm an existing file as an example to how you want it to produce the output and tell it "Do X in the style of Y.file". Don't forget with the latest models and huge context windows you could very well provide entire subdirectories into context (although I would recommend being pretty targeted still)
- the model/tool you're using sucks
- you work in a problem domain that LLMs are genuinely bad at
see here for Cline team's perspective on this topic: https://www.reddit.com/r/ChatGPTCoding/comments/1kymhkt/clin...
I suggest this as the bare minimum for the HN community when discussing their bad results with LLMs and coding:
- what is your problem domain
- show us your favorite prompt
- what model and tools are you using?
- are you using it as a chat or an agent?
- are you bringing your own keys or using a service?
- what did you supply in context when you got the bad result?
- how did you supply context? copy paste? file locations? attachments?
- what prompt did you use when you got the bad result?
Most of the cynics would not provide even half of this because it'd be embarrassing and reveal that they have no idea what they are talking about.
> Previously I would have had to have 1 million dollars sitting around to fund the amount of output that I am able to produce with a $20 subscription to an LLM service.
this sounds like the death of employment and the start of plutocracy
not what I would call "democratisation"
Well, I've said enough about cynicism here so not much else I can offer you. Good luck with that! Didn't realize everybody loved being an employee so much
so, employee or destitute? tough choice