5 ms·

My feeling is: Who cares? What matters is if the text is factual. Humans without AI can lie and mislead as well. If ChatGPT and other tools help humans write

My feeling is: Who cares? What matters is if the text is factual. Humans without AI can lie and mislead as well. If ChatGPT and other tools help humans write nice, easy to read text from prompts, more power to them. Except for professors trying to grade assignments, the average person should not care. I think this mostly affects a certain educated person who gate-keeps around writing skill and is upset that the unwashed masses can now write like them.

Veen 4y ago
It matters because LLMs can tell plausible lies at incredible scale: marketing, propaganda, misinformation and disinformation, etc. Understanding whether content is AI generated would be a useful red flag, but we can't. Nor can supposed "AI detectors" do so with any reliability [0]. It's going to be a problem. [0]: https://arxiv.org/abs/2303.11156 https://arxiv.org/abs/2303.11156
- callahad 4y ago
  It took me a few weeks, but I've landed firmly in the existential despair camp. Within a year, the noise floor will have shot through the roof, and I'm not sure how we'll winnow truth from weaponized, hyperscale hallucinations. Maybe the good news is that the problem will likely arrive so quickly that by the time we're done collectively comprehending the ways in which it could play out, it will have. And then we can dispense with the hypotheticals and get on with the work of clawing back a space for humans.
  - CatWChainsaw 4y ago
    Welcome to the doom camp :)
macNchz 4y ago
For one it’s an absolutely massive force multiplier for scammers who often do not write well in English, and who have so far been constrained by human limits in how many victims they can have “in process” at once.
- Joker_vD 4y ago
  The "cold-call" spam letters have to be written in poor English because spammers want only gullible enough people to respond to them because, as you've said, they're constrained in how many marks they can process simultaneously. So they arrange this self-selection process where too sceptical people bail out as early as possible at as small as possible cost for the scammers.
nonethewiser 4y ago
> I think this mostly affects a certain educated person who gate-keeps around writing skill and is upset that the unwashed masses can now write like them. Unwashed masses can’t write like then though. A few AIs can. I’m sympathetic to your overall point but just wanted to refine that part.

Veen 4y ago

It matters because LLMs can tell plausible lies at incredible scale: marketing, propaganda, misinformation and disinformation, etc. Understanding whether content is AI generated would be a useful red flag, but we can't. Nor can supposed "AI detectors" do so with any reliability [0]. It's going to be a problem. [0]: https://arxiv.org/abs/2303.11156 https://arxiv.org/abs/2303.11156

callahad 4y ago
It took me a few weeks, but I've landed firmly in the existential despair camp. Within a year, the noise floor will have shot through the roof, and I'm not sure how we'll winnow truth from weaponized, hyperscale hallucinations. Maybe the good news is that the problem will likely arrive so quickly that by the time we're done collectively comprehending the ways in which it could play out, it will have. And then we can dispense with the hypotheticals and get on with the work of clawing back a space for humans.
- CatWChainsaw 4y ago
  Welcome to the doom camp :)

macNchz 4y ago

For one it’s an absolutely massive force multiplier for scammers who often do not write well in English, and who have so far been constrained by human limits in how many victims they can have “in process” at once.

Joker_vD 4y ago
The "cold-call" spam letters have to be written in poor English because spammers want only gullible enough people to respond to them because, as you've said, they're constrained in how many marks they can process simultaneously. So they arrange this self-selection process where too sceptical people bail out as early as possible at as small as possible cost for the scammers.

nonethewiser 4y ago

> I think this mostly affects a certain educated person who gate-keeps around writing skill and is upset that the unwashed masses can now write like them. Unwashed masses can’t write like then though. A few AIs can. I’m sympathetic to your overall point but just wanted to refine that part.