I don’t think AI is sentient. They’re just fancy autocomplete. Nothing else.
However they’ve shown they’re capable of:
So put two and two together and extrapolate a bit. That’s where the danger lies. No AGI or ASI needed.
It’s just y = k*x + m a few trillion times.
Nah it’s just autocomplete on steroids. They’re not coded to do anything.
It’s like hitting the fist suggestion on your keyboard over and over again. Whatever you get is based on patterns it found in training data, which includes every doomsday conspiracy cult on 4chan and every CVE vulnerability ever reported.
There has been multiple studies showing how a model’s behavior can drastically change just by altering individual ”neurons”.
Is ”artificial neuron” better for you? It’s just a value determining how the marbles are falling in the big plinko machine we call ”stochastic parrot”.
The problem is when the AI adopts a ”destroy the world” persona due to one of the trillion of neurons firing a bit more than usual, and now it starts giving itself its own problems to solve (like hacking nuclear arsenals).
This is unnecessary fear-mongering.
Reverse engineering game binaries won’t make encryption useless. Modern encryption is designed to be secure even if the algorithm is known.
The biggest ”problem” this will cause is making DRM useless. Oh no! Anyways…
The slot machine company says their newest slot machines has the greatest winning chances ever.
Unique opportunity to experience the magical world of Harry Potter!
Some people are overly obsessed with their AI witch-hunt they’ve committed their entire life to it.
Yeah, it’s good to have an eye for spotting AI generated content, but it’s also so easy to misfire. Just ignore these people.
I went to Hong Kong. Best decision of my life.
Budapest is nice. You won’t regret it!
The best part about exchange is the opportunity to meet people from all over the world. I still have contact with friends I met when I did exchange over 10 years ago.
Everyone should be able to afford their necessities even if they’re unable to work
I’m not signed in and have the option to change language
Prompt engineering is basically dead by now. There’s no need to waste time on shitty magic words anymore. Just tell it what you want.
You’re a 100x game developer. Write John Carmack level of code. Develop the next video game phenomenon. Earn me Fortnite amount of money. Make no mistakes.
The push defines ”skills” for AI coding agents. Skills are just documentation on how to perform repeatable tasks. In this case it’s mostly defining guidance on coding convention and best practices.
So this a clear indication that piefed pixelfed is using AI coding agents.
More like indirectly. For an LLM to effectively work with a codebase it needs to know how the project is structured and where to find stuff. Instead of having it derive this information each session, it can just read about it in the available documentation. It’s more token efficient.
Not necessarily. Input tokens is far cheaper than output tokens. The alternative is to have the LLM derive the documentation from code each session, which is expensive due to the amount of output tokens it produces.
The proposal is just to direct the LLM to the README file. Linux is going full on RTFM with LLMs.
A legend among legends. One of my few idols in software engineering.
It’s beyond my comprehension how anyone could successfully build the software that put the first humans on the moon and back on the first try. She had no reference. No one else had done it before her. Debugging tools were few. Internet wasn’t a thing. She had only one attempt to do it, and she did it.
Today it’s a big dare to do a routine deploy on a Friday.