“ll” — Fediverse search results on home.social

mnl mnl mnl mnl mnl @mnl · 2026-05-14 · 22:12 UTC

dude i don't know what it is, but how does every single article about "AI is going to rot our brain" manage to link papers that sometimes barely make a point or in this case seem to be completely unrelated to the topic?

https://futurism.com/artificial-intelligence/ai-college-students-homogenized

Like is this just lazy non-AI writing? Because legit even chatgpt or claude on normal thinking do better.

#llms #llm #genai

saxnot ➡️ GPN, RockHarz @[email protected] · 2026-05-14 · 17:13 UTC

One does not understand code copied from the #LLM. It's the same as pasting from a textbook or from colleague at uni: you THINK you understand it but you don't.

We fool ourself and say "yeah I *COULD* write it on my own"

Copying the solution to a math problem from someone who solved it is dangerous because "yeah I understand it" is not "I learned this to do it myself".
That's why learning groups with diverse skill level are so dangerous.

#llm

saxnot ➡️ GPN, RockHarz @[email protected] · 2026-05-14 · 17:13 UTC

One does not understand code copied from the #LLM. It's the same as pasting from a textbook or from colleague at uni: you THINK you understand it but you don't.

We fool ourself and say "yeah I *COULD* write it on my own"

Copying the solution to a math problem from someone who solved it is dangerous because "yeah I understand it" is not "I learned this to do it myself".
That's why learning groups with diverse skill level are so dangerous.

#llm

saxnot ➡️ GPN, RockHarz @[email protected] · 2026-05-14 · 17:13 UTC

One does not understand code copied from the #LLM. It's the same as pasting from a textbook or from colleague at uni: you THINK you understand it but you don't.

We fool ourself and say "yeah I *COULD* write it on my own"

Copying the solution to a math problem from someone who solved it is dangerous because "yeah I understand it" is not "I learned this to do it myself".
That's why learning groups with diverse skill level are so dangerous.

#llm

saxnot ➡️ GPN, RockHarz @[email protected] · 2026-05-14 · 17:13 UTC

One does not understand code copied from the #LLM. It's the same as pasting from a textbook or from colleague at uni: you THINK you understand it but you don't.

We fool ourself and say "yeah I *COULD* write it on my own"

Copying the solution to a math problem from someone who solved it is dangerous because "yeah I understand it" is not "I learned this to do it myself".
That's why learning groups with diverse skill level are so dangerous.

#llm

saxnot ➡️ GPN, RockHarz @[email protected] · 2026-05-14 · 17:13 UTC

One does not understand code copied from the #LLM. It's the same as pasting from a textbook or from colleague at uni: you THINK you understand it but you don't.

We fool ourself and say "yeah I *COULD* write it on my own"

Copying the solution to a math problem from someone who solved it is dangerous because "yeah I understand it" is not "I learned this to do it myself".
That's why learning groups with diverse skill level are so dangerous.

#llm

Raku updated modules @raku_updated_modules · 2026-05-14 · 17:05 UTC

LLM::Prompts 0.2.15
Facilitating the creation, storage, retrieval, and curation of LLM prompts.
https://raku.land/zef:antononcube/LLM::Prompts

datum (n=1) @[email protected] · 2026-05-14 · 16:59 UTC

LLMs have been slopcoded to not talk about Goblins.

Should webpages have "prove you are human by writing a paragraph about goblins" instead of nasty captchas?

#LLM #LLMs #AI #Goblins #Goblin

#llm #llms #ai #goblins #goblin

datum (n=1) @[email protected] · 2026-05-14 · 16:59 UTC

LLMs have been slopcoded to not talk about Goblins.

Should webpages have "prove you are human by writing a paragraph about goblins" instead of nasty captchas?

#LLM #LLMs #AI #Goblins #Goblin

#llm #llms #ai #goblins #goblin

datum (n=1) @[email protected] · 2026-05-14 · 16:59 UTC

LLMs have been slopcoded to not talk about Goblins.

Should webpages have "prove you are human by writing a paragraph about goblins" instead of nasty captchas?

#LLM #LLMs #AI #Goblins #Goblin

#llm #llms #ai #goblins #goblin

datum (n=1) @[email protected] · 2026-05-14 · 16:59 UTC

LLMs have been slopcoded to not talk about Goblins.

Should webpages have "prove you are human by writing a paragraph about goblins" instead of nasty captchas?

#LLM #LLMs #AI #Goblins #Goblin

#goblin #goblins #ai #llms #llm

datum (n=1) @[email protected] · 2026-05-14 · 16:59 UTC

LLMs have been slopcoded to not talk about Goblins.

Should webpages have "prove you are human by writing a paragraph about goblins" instead of nasty captchas?

#LLM #LLMs #AI #Goblins #Goblin

#llm #llms #ai #goblins #goblin

PaulaToThePeople 😷 @[email protected] · 2026-05-14 · 16:49 UTC

I feel like we're living in a time where 80% of the population of the "western" world (lapa countries) have a very energy-intensive computerchip implanted in their brains. It runs on fossil fuels that the people have to drink or eat. It gives them electro shocks that kill brain cells and what is the function? Whenever two people with such a chip get close to each other they get a shock that sends info through their nerves to their muscles to make a nazi salute.
And people find it so convenient that they no longer have to put effort into greeting people or any human interactions at all and can just stare at their phones all day.

#AI #NoAI #LLMMeAlone

#llmmealone #noai #ai