__loam

46 ms·

751.
by __loam 2y ago
What's the point without the merlin engine?
752.
by __loam 2y ago
When will programmers get it through their thick skulls that an artist taking inspiration from a style and a well funded tech corporation downloading 400 million images to train on are two different things that shouldn't be compared. G
753.
by __loam 2y ago
Using the product of someone's labor to build for profit systems that are trying to put them out of work remains fucked up. And no, a simple copyright filter isn't the problem. The problem is that the work was used without permiss
754.
by __loam 2y ago
I just want to buy a screen guys.
755.
by __loam 2y ago
Hope we can turn this off
756.
by __loam 2y ago
I agree from the perspective that they're probably using the data to do training which is a big big no no in the space, but in terms of hacking, that's true of basically any platform. Certifications like SOCII and Fedramp exist, a
757.
by __loam 2y ago
Companies continuing the pattern of offloading training costs while complaining that it's hard to find skilled workers. LLMs are just an excuse.
758.
by __loam 2y ago
Problems like auto-generating spam emails for Nigerian princes.
759.
by __loam 2y ago
I would be interested in how specialist companies like Harvey, which is backed by OpenAI, are doing in this space. They're pretty secretive though.
760.
by __loam 2y ago
Sounds like an awesome way for an entire professional class to slowly lose their skills.
761.
by __loam 2y ago
What a sad view of things.
762.
by __loam 2y ago
An interaction like this led me to cancel my chatgpt subscription.
763.
by __loam 2y ago
It's a quality signal. Not all my teachers were good and the system is underfunded, but most of turn cared about the students and I'm incredibly thankful for the education they gave me. I trust the opinions of people teaching kids
764.
by __loam 2y ago
Also incredible levels of Dunning Kruger effect for fields outside of programming like education.
765.
by __loam 2y ago
As someone who also went through the public school system and who was able to get into an ivy league school off of that education, maybe you should show some more respect to the people doing the work of education for shit pay.
766.
by __loam 2y ago
I hate listening to tech people talk about education like it's some problem to be solved. Using a system that you know is wrong 10% of the time for education is a terrible idea. We're trying to teach things like basic literacy and
767.
by __loam 2y ago
> best Lmao
768.
by __loam 2y ago
https://teachyourselfcs.com/
769.
by __loam 2y ago
In addition to what everyone else has said, pixel art relies a lot of techniques like AA, dithering, outlines, specific ways of drawing curves, limited palettes, and implied detail that make it difficult to just convert from a vector piece.
770.
by __loam 2y ago
Graphql was routinely an overly complex piece of shit when I was forced to deploy software with it.
771.
by __loam 2y ago
Someone post the jungle quote, I'm tired.
772.
by __loam 2y ago
I've seen a lot of salivation over replacing customer service reps lol.
773.
by __loam 2y ago
Not sure OpenAI wants to invite the negative press at a time when the public opinion on them is wavering.
774.
by __loam 2y ago
You probably also need that for the AI as well though
775.
by __loam 2y ago
I think the problem here is that that is often still not that acceptable. Let's imagine a system with say, 100 million users making 25 queries a day, just to give us some contrived numbers to examine. At a 10% error rate that's 25
776.
by __loam 2y ago
This seems like it could do some cool code completion stuff with local models.
777.
by __loam 2y ago
Yeah that sounds way better. I saw one of the python libraries they recommended mention retries and I thought, this can't be that awful can it?
778.
by __loam 2y ago
Google is so terrified that someone is threatening their market position, the one in which they have over $100b in cash and get something like $20b in profit quarterly, that they're willing to shove this technology into some of the mos
779.
by __loam 2y ago
I feel like an insane person everytime I look at the LLM development space and see what the state of the art is. If I'm understanding this correctly, the standard way to get structured output seems to be to retry the query until the st
780.
by __loam 2y ago
I think what happened is OpenAI was probably going to easily defend themselves from this suit until Altman tweeted about it. It seems like the company didn't really do anything wrong (in this case) but that tweet is actually going to g

More ›