4 ms·

These AI detectors have all the validity of phrenology. You can enter some of your own writing (I did) to easily see for yourself. The text I entered had been o

by hilux 3y ago

These AI detectors have all the validity of phrenology. You can enter some of your own writing (I did) to easily see for yourself. The text I entered had been online, and still is online, longer than AI writers have been around - but that doesn't seem to feed into their logic.

CamperBob2 3y ago
The whole idea of an "AI detector" is stupid. If it ever does work, it will only work until someone trains a model against it.
- bee_rider 3y ago
  And anyway, text exists to be read by humans. If the buyer needs an “AI detector” to tell if the copy was AI generated, the AI would seem to be good enough. I think this only exists to facilitate some already-bad heuristics, like the idea that nobody in-house needs to check the contractor’s work-product.
ceejayoz 3y ago
A good example: Put bits of the US Constitution in some AI detectors and they'll tell you it was AI generated. https://arstechnica.com/information-technology/2023/07/why-ai-detectors-think-the-us-constitution-was-written-by-ai/ https://arstechnica.com/information-technology/2023/07/why-a...
- lemmsjid 3y ago
  Interesting--perhaps the AI detectors are right, and this will be how we discover that we are in fact in a simulated universe!
  - ElevenLathe 3y ago
    ...or at least a simulated democracy.
RationalDino 3y ago
Which text did you enter, and into which AI detector? I tried it with my blog, and it was correctly identified by Copy Leaks as written by a human. I don't know how representative https://copyleaks.com/blog/ai-detector-continues-top-accuracy-third-party https://copyleaks.com/blog/ai-detector-continues-top-accurac... is, but certainly all of them are getting above phrenology detection rates. Even with this guy, the problem wasn't that it misidentified him as AI. It was that "we can't tell" got turned into, "we can't risk Google thinking he might be AI".
- pixl97 3y ago
  ""we can't tell" got turned into, "we can't risk" Oh god, we've developed yet another shitty lie detector.
  - RationalDino 3y ago
    You only just noticed that?

by hilux 3y ago

CamperBob2 3y ago

The whole idea of an "AI detector" is stupid. If it ever does work, it will only work until someone trains a model against it.

bee_rider 3y ago
And anyway, text exists to be read by humans. If the buyer needs an “AI detector” to tell if the copy was AI generated, the AI would seem to be good enough. I think this only exists to facilitate some already-bad heuristics, like the idea that nobody in-house needs to check the contractor’s work-product.

ceejayoz 3y ago

A good example: Put bits of the US Constitution in some AI detectors and they'll tell you it was AI generated. https://arstechnica.com/information-technology/2023/07/why-ai-detectors-think-the-us-constitution-was-written-by-ai/ https://arstechnica.com/information-technology/2023/07/why-a...

lemmsjid 3y ago
Interesting--perhaps the AI detectors are right, and this will be how we discover that we are in fact in a simulated universe!
- ElevenLathe 3y ago
  ...or at least a simulated democracy.

RationalDino 3y ago

Which text did you enter, and into which AI detector? I tried it with my blog, and it was correctly identified by Copy Leaks as written by a human. I don't know how representative https://copyleaks.com/blog/ai-detector-continues-top-accuracy-third-party https://copyleaks.com/blog/ai-detector-continues-top-accurac... is, but certainly all of them are getting above phrenology detection rates. Even with this guy, the problem wasn't that it misidentified him as AI. It was that "we can't tell" got turned into, "we can't risk Google thinking he might be AI".

pixl97 3y ago
""we can't tell" got turned into, "we can't risk" Oh god, we've developed yet another shitty lie detector.
- RationalDino 3y ago
  You only just noticed that?