Live data from Hacker News

"No, I swear I wrote this."

revise.io

1–10 of 50 posts

Re: "No, I swear I wrote this."

#3
post #2

> Boom, definitive "proof of typing" that a given piece was produced by genuine meat-on-keyboard effort. Except it lacks proof of the keyboard - and the meat.

Yep, someone can "vibe code" something, that would emulate that human, the typing, the pauses, random typos, backspaces, and edits. There are probably models already available that describe the average delays between two consecutive keypresses depending on the location of keys, etc.

Re: "No, I swear I wrote this."

#5
post #2

> Boom, definitive "proof of typing" that a given piece was produced by genuine meat-on-keyboard effort. Except it lacks proof of the keyboard - and the meat.

Yep, someone can "vibe code" something, that would emulate that human, the typing, the pauses, random typos, backspaces, and edits. There are probably models already available that describe the average delays between two consecutive keypresses depending on the location of keys, etc.

I'm pretty sure Coursera or edX was using this same approach a while back — they'd make you type in a paragraph of text, then they'd use the timing information as a fingerprint or signature, to authenticate you as the actual student.

Re: "No, I swear I wrote this."

#6
post #4

What if the human is just transcribing directly from AI generated text?

One could, but the primary motivation to use AI to generate text is because it is fast and easy. You could spend an hour elaborately pretending to write something yourself, or you could simply write it yourself.

Re: "No, I swear I wrote this."

#8
Trivial to simulate. Though I disagree that AI writing is impossible to differentiate from human prose, at least for now. It's still pretty obvious and still much worse than human prose (or, at least, much less interesting to read), though some are better than others, and I'm better able to spot writing from models I use regularly (Claude has a very distinctive style I can spot from a mile away, but that's true partly because I read its prose nearly every day when using it for coding).

Re: "No, I swear I wrote this."

#9
post #2

> Boom, definitive "proof of typing" that a given piece was produced by genuine meat-on-keyboard effort. Except it lacks proof of the keyboard - and the meat.

Yep, someone can "vibe code" something, that would emulate that human, the typing, the pauses, random typos, backspaces, and edits. There are probably models already available that describe the average delays between two consecutive keypresses depending on the location of keys, etc.

surely there real-time AI writing checkers? predictability in word choices and sentence-length variation are alr available so maybe someone has to make something that measures delays in sec now. that could be a feature that op implements *like coursera/edx tech as sidpatil mentions

Re: "No, I swear I wrote this."

#10
post #8

Trivial to simulate. Though I disagree that AI writing is impossible to differentiate from human prose, at least for now. It's still pretty obvious and still much worse than human prose (or, at least, much less interesting to read), though some are better than others, and I'm better able to spot writing from models I use regularly (Claude has a very distinctive style I can spot from a mile away, but that's true partl…

Depends on what’s being written, and who the audience is. Anything of any length would be hard to simulate in a way that would fool an author - writing has a certain flow to it. A cadence. The editing and restructuring, deleting of words, typos you don’t catch until some random reread, rephrasing of sections because you want to use the original phrasing later in the piece.

Could you simulate something be typed? Trivially. Could you simulate something be drafted? Honestly, even if you wanted to put in all that time and effort, I’m not even sure LLMs are sophisticated enough to send the logical drafts, loops and edits that would pass a writers sniff test

Post reply on HN