Artisan Loaf 🏳️‍🌈
artisanloaf.bsky.social
Artisan Loaf 🏳️‍🌈
@artisanloaf.bsky.social
mostly very tired

💜🏳️‍🌈🏳️‍⚧️💜
I mean it even applies to good creative works, TV shows and movies are written and edited by groups, and are not one person sitting at a typewriter.

So if you are not reading books, long form pieces/blogs etc.... you probably never read one person's writing.
November 17, 2025 at 9:13 PM
Just pondering here, but so much of what people see/hear/read isn't written by a person anyway, it's designed/written/edited by teams of people.

Half the time i'm writing corp/biz emails, I feel like i could be an llm.
November 17, 2025 at 9:09 PM
I am coming to terms with the reality that Toyota is just not it anymore.
November 17, 2025 at 8:30 PM
as a local ai endorser... economies of scale [shocked pikachu]
November 17, 2025 at 8:26 PM
do they have a merch store?
November 17, 2025 at 11:43 AM
Reposted by Artisan Loaf 🏳️‍🌈
I'm long done persuading anyone because honestly at this point if you got your empathy surgically removed there's nothing to discuss
November 17, 2025 at 9:27 AM
it has less crumbs/sauce on the back.
November 15, 2025 at 8:48 PM
whoa!
November 15, 2025 at 8:35 PM
I had the Ford version at one point, and thought they were solid trucks.
November 15, 2025 at 8:33 PM
sometimes i write a bunch of notes and crap, and then get an LLM to organise/structure it better for me, without changing any of the substance. especially helpful with late night epiphanies that probably wont make much sense in the morning if an LLM didn't edit me a bit.
November 15, 2025 at 7:42 PM
Point is, if either or both go under, the technology still exists and is useful.

Despite what some commentators imply, the LLM/GenAI world exists outside of trillion parameter proprietary models by unprofitable unicorns.
November 15, 2025 at 7:35 PM
It sure as fuck does.

As i've said before....

The dot-com bubble was real, those companies were worthless relative to their valuations, and deserved to go under... And that was the last time anyone used the internet.
November 15, 2025 at 7:22 PM
If OpenAI and Anthropic went under tomorrow.

We still have Nvidia hardware and open weight Qwen/Llama/Mistral/Gemma etc etc etc.

The genie is out of the bottle, even if Sam Altman is a fraud.
November 15, 2025 at 7:19 PM
I know I could get more out of it than I do, as the technology currently exists, plenty of room for me to refine and expand my own uses without any sota advances.
November 15, 2025 at 7:16 PM
gpt4.1-mini with tool use and image input actually incredibly solid and cheap, avg 8seconds and $0.001 for each asset logged in the database with all the info needed in the right places.
November 15, 2025 at 7:11 PM
It's a tool that takes time, practice and a good understanding of the limitations to use properly.

WalMart will sell anyone a cordless drill, chainsaw etc... The instructions on the box don't actually tell you how to use it 'properly'... They are dangerous and destructive when used poorly.
November 15, 2025 at 7:03 PM
I've made many internal tools that do exactly what we need and 'just work'... some use llms, many were just built with llms... sometimes i just need a python script to do one thing once, and an llm can put that together based on my requirements in a minute usually, and then i'm on to the next thing.
November 15, 2025 at 6:54 PM
Still think some people on this website are shocked that not everyone is a writer or artsy creative.

I spent all day using LLMs for asset tagging and inventory management. It saves me and other people manually entering every make, model, serial, mac address for thousands of devices.
November 15, 2025 at 6:36 PM
Trying to square the circle with how the elite liberal deep state has been using systems to destroy america and make life awful for those with a drive to succeed, or whatever bullshit he says in any other situation.
November 13, 2025 at 1:42 PM