3 ms·

One of my favorite LLM-smells to train myself to sniff out is how well the author can tie two seemingly disparate ideas together into something compelling or no

One of my favorite LLM-smells to train myself to sniff out is how well the author can tie two seemingly disparate ideas together into something compelling or novel. The more ambitious the synthesis and the less impressed I am the more likely I am to assume it was at least partially the work of an LLM. Previously, "Jimi Hendrix Was a Systems Engineer" <https://spectrum.ieee.org/jimi-hendrix-systems-engineer https://spectrum.ieee.org/jimi-hendrix-systems-engineer>

shevy-java 1mo ago
The content was clearly AI written. It made no sense. Even crazy humans such as Terry Davis made more sense than such AI slop generated "blog" entries.
- tolerance 1mo ago
  That this is published by the lead of a pretty popular piece of software is disconcerting. The UI for the IDE itself and the web design for Delta even have that Claude Code "Claudhaus" look to them. I'm beginning to think that this sort of behavior is actually an inverted counter-signal. The average person attuned to LLM's markings may find these things distasteful. But certain individuals or organizations may feel right at ease seeing them.
  - segfaltnh 1mo ago
    And funding them.
patcon 1mo ago
Ah, so like the distance between ambition and impact in metaphor/synthesis? I like that.
- tolerance 1mo ago
  Exactly.
- more-coffee 1mo ago
  Do you mean that LLMs are unable to use metaphors effectively? (Just clarifying because English is not my first language)
  - patcon 1mo ago
    heh, yeah, just "unable to use metaphor effectively" is what I meant :) i guess I'm saying it in a funny way because my brain has always thought about semantic space (and the act/object of conversation itself) in a very spatial way, and it seems that LLMs have a similar tendency to elaborate: "unable to use metaphor effectively" seems to have multiple dimensions of failure, in which I find the distance metaphor to make easier to explore. A mind-like thing can (1) not know how to make the leaps (and so never try to make them), or (2) it can leap too ambititously and land in faraway places that don't have any deep symmetry in meaning... it can (3) also maybe "not stick the landing" while jumping someplace that theoretically might have merit (in a sense of global vs local scale, maybe being ok at the global scale orienting, but not local scale)... So in my defence, I think there are failure modes of "being ineffective at metaphor" that are masked by speaking in plainer language :)

shevy-java 1mo ago

The content was clearly AI written. It made no sense. Even crazy humans such as Terry Davis made more sense than such AI slop generated "blog" entries.

tolerance 1mo ago
That this is published by the lead of a pretty popular piece of software is disconcerting. The UI for the IDE itself and the web design for Delta even have that Claude Code "Claudhaus" look to them. I'm beginning to think that this sort of behavior is actually an inverted counter-signal. The average person attuned to LLM's markings may find these things distasteful. But certain individuals or organizations may feel right at ease seeing them.
- segfaltnh 1mo ago
  And funding them.

patcon 1mo ago

Ah, so like the distance between ambition and impact in metaphor/synthesis? I like that.

tolerance 1mo ago
Exactly.
more-coffee 1mo ago
Do you mean that LLMs are unable to use metaphors effectively? (Just clarifying because English is not my first language)
- patcon 1mo ago
  heh, yeah, just "unable to use metaphor effectively" is what I meant :) i guess I'm saying it in a funny way because my brain has always thought about semantic space (and the act/object of conversation itself) in a very spatial way, and it seems that LLMs have a similar tendency to elaborate: "unable to use metaphor effectively" seems to have multiple dimensions of failure, in which I find the distance metaphor to make easier to explore. A mind-like thing can (1) not know how to make the leaps (and so never try to make them), or (2) it can leap too ambititously and land in faraway places that don't have any deep symmetry in meaning... it can (3) also maybe "not stick the landing" while jumping someplace that theoretically might have merit (in a sense of global vs local scale, maybe being ok at the global scale orienting, but not local scale)... So in my defence, I think there are failure modes of "being ineffective at metaphor" that are masked by speaking in plainer language :)