I think the current issue is if you train something to write code and poetry and recipes etc. it's going to reduce your accuracy at each task.

But as you say we'll see.

Like most automations they'll never replace us fully. But we're already seeing productivity boosts that will mean not as many people are needed.

Kind of like how excel massively reduced how many accountants you need.

LLMs are general purpose, basically for demonstration purposes. This is basically generation 0.

Once they can optimize the training process more, which is something being heavily researched, you'll be able to create ones for specific languages or even frameworks.

Dedicated hardware will be another huge boost.

Then they'll start to be amazing.

Likely aimed at financial departments which often (in my experience) love to do everything in Excel.

I feel people who complain about LLMs really don't use them correctly.

I use them a lot at work. But generally as a learning tool or interactive docs rather than a code generator.

I've found them to be incredibly useful.

All the use after free and buffer overflow bugs that plague our key infrastructure.

This guy needs to be in fucking prison. What is it going to take. Fuck.

Not sure reading comprehension is tied to typos on a mobile. I was referring to the title of the post. No need to be rude friend.

This is obviously fucked up. But they are clearly intentionally downplaying that he pulled a knife when stopped. He wasn't shot for the fair lol

Absolutely. But you are still downplaying the amount of prep a DM has to do to make it playable and fleshed out.

In the UK unless your parents are particularly poor it is not that common to support them.

We have a socialised pension that most should be able to live on. And most people have private or government pensions as well.

I think that statement massively downplays the amount of work a DM puts in behind the scenes to create the world.

I was talking more generally about LLMs here

I've found they're great as a learning tool where decent docs are available. Or as an interactive docs you can ask follow up questions to.

We mostly use c# and it's amazing at digging into the MS docs to pull out useful things from the bcl or common patterns.

Our new juniors got up to speed so fast by asking it to explain stuff in the existing codebases. Which in turn takes pressure off more senior staff.

I got productive in vuejs in a large codebase in a couple days that way.

Using to generate actual code is insanely shit haha It is very similar to just copy pasting code and hacking it in without understanding it.

That's my thinking. If this is the best they can come up with there must not be a lot going on.

Time to learn about context

I think with http request failures losing the status code can be a real pain. There's a big difference between 400, 401, 404 and 500 for instance.

I wouldn't call bad readability a loaded gun really. Your dev tools will hopefully make it pretty easy to learn the type. It should be a minor inconvenience at best.

That is in fact literally what it means and what we are talking about here.

I'm not and stated that in my first comment.