There seem to be two distinct kinds of AI Psychosis occuring:
Type 1: Becoming delusional about themselves and the world as a result of being gasslit or glazed by the "friend in their computer" thinking they’re inventing new physics, discovering mystical secrets or generating megabytes of useless output;
Type 2: Becoming delusional about what LLMs are capable of and how effective they are, as a result of developing a reliance upon them, and becoming fanatical in their promotion and defense.
Type 2.5: Such Jerks who insist on policing other people's use of LLM's as if it's the most egregious crime ever. It's not.
It may be legally and even morally dubious; to be avoided, minimised and deprecated but no-one appointed you (or me) as the AI/LLM Police. If you want to use such products then carry on. If you don't that's fine also. Being a jerk about it will get you blocked.
If you actually don't vibe code then you cannot actually have imposter syndrome. Just look at the code quality from someone who is an actual imposter at programming with vibes.
LLMs, by their nature, generate statistically plausible output. That is often an output set that overlaps with the correct output. However outputs that are not correct look exactly the same as the ones that are.
You're reading the statistical output of a very large matrix multiplication and assuming it makes sense. It does not. By design.
AI is just a tool, like a calculator.
If calculators were rented, unreliable, dependent on endless data centers, monopolizing the market for electronic components, responsible for mass copyright violation, and shredding rare books.
- Humans must not anthropomorphise AI systems,
- Humans must not blindly trust the output of AI systems,
- Humans must remain fully responsible and accountable for consequences arising from the use of AI systems.
#NoAI
#FriendsOfEliza
#PPQS
#AIPsychosis