Woops. Sorry mods. Reposting with links.
https://nitter.space/thelillygaddis/status/1904852790460965206#m
this post was submitted on 27 Mar 2025
1163 points (99.0% liked)
People Twitter
6793 readers
1420 users here now
People tweeting stuff. We allow tweets from anyone.
RULES:
- Mark NSFW content.
- No doxxing people.
- Must be a pic of the tweet or similar. No direct links to the tweet.
- No bullying or international politcs
- Be excellent to each other.
- Provide an archived link to the tweet (or similar) being shown if it's a major figure or a politician.
founded 2 years ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
Any AI model is technically a black box. There isn't a "human readable" interpretation of the function.
The data going in, the training algorithm, the encode/decode, that's all available.
But the model is nonsensical.
That’s not true, there are a ton of observabity tools for the internal workings.
The top post on HN is literally a new white paper about this.
https://news.ycombinator.com/item?id=43495617
Thank you that's amazing
They also made a video:
https://youtu.be/Bj9BD2D3DzA
Some simpler "AI models" are also directly explainable or readable by humans.
In almost exactly the same sense as our own brains' neural networks are nonsensical :D
Yeah despite the very different evolutionary paths there's remarkable similarities between idk octopus/crow/dolphin cognition