News for hackers. Stuff that matters.
Re:Working on Economics with Fable 5 by owenpayton on Monday September 07, 2026 @07:15PM (#49604014)
Re:Working on Economics with Fable 5 by estearum on Monday September 07, 2026 @07:19PM (#49604030)
It may well be, but also Pangram is ridiculously transparent pseudoscience. Let's not play pretend that it's authoritative, please.

Edit: Just for fun I just defeated it with AI-generated text that it reported as 100% human-written. Great stuff. Truly genius scam.

Re:Working on Economics with Fable 5 by ekelsen on Monday September 07, 2026 @07:26PM (#49604085)
I never found a piece of text that pangram claims is 100% AI that to me seems like purely human written text.

The only counter example is when a human composed a piece deliberately trying to mimic AI slop.

So call it an AI slop detector if you want, but either way it accurately identifies stuff I don't wan't to read.

Re:Working on Economics with Fable 5 by estearum on Monday September 07, 2026 @07:28PM (#49604096)
> I never found a piece of text that pangram claims is 100% AI that to me seems like purely human written text.

Is that the level of sensitivity we're looking for? More than literally 0%? Do you know how crazy it'd be to achieve that, even if they were trying to?

Re:Working on Economics with Fable 5 by porridgeraisin on Monday September 07, 2026 @07:25PM (#49604077)
It is not pseudoscience. Have you read about how it works?
Re:Working on Economics with Fable 5 by estearum on Monday September 07, 2026 @07:32PM (#49604123)
Yes, quite a bit actually. It's directly analogous to e.g. "Red sky at night, sailors' delight", i.e. a pseudo-scientific approach that probably has a better-than-random success rate.
Re:Working on Economics with Fable 5 by robotresearcher on Monday September 07, 2026 @08:12PM (#49604347)
So it’s a heuristic. Lots of very valuable heuristics exist.
Re:Working on Economics with Fable 5 by estearum on Monday September 07, 2026 @08:51PM (#49604480)
None of which is authoritative, per my initial comment.
Re:Working on Economics with Fable 5 by Tostino on Monday September 07, 2026 @10:27PM (#49605066)
And I'd agree if that's what GP said.
Re:Working on Economics with Fable 5 by robotresearcher on Tuesday September 08, 2026 @02:18AM (#49606173)
What name do you use for an approach that probably gets a better than random success rate (without guaranteeing that)?

I’m leaving out the ‘pseudo-scientific’ part of GP’s description, since it doesn’t entail anything technically.

Re:Working on Economics with Fable 5 by Tostino on Tuesday September 08, 2026 @06:01AM (#49608132)
I was talking about the root comment that just said "it's AI because this website said so", not the one about treating it like a heuristic. I agree with that.
Re:Working on Economics with Fable 5 by snitty on Monday September 07, 2026 @08:33PM (#49604373)
What were the text, your prompt, and model?
Re:Working on Economics with Fable 5 by estearum on Monday September 07, 2026 @08:51PM (#49604476)
It was actual work output I generated a few weeks back with Sonnet 5. Not posting here for obvious reasons, but it was nothing special or engineered for Pangram deception.
Re:Working on Economics with Fable 5 by ekelsen on Monday September 07, 2026 @10:49PM (#49605189)
The real reason being that it doesn't exist?
Re:Working on Economics with Fable 5 by estearum on Tuesday September 08, 2026 @08:32AM (#49609477)
Ya bro it's totally not possible to defeat a detection system that, in its own marketing on contrived, game-able benchmarks that they created, claims only a 99.6% true negative rate.

It has literally never been done.

What a hilarious inversion of credulity.

Re:Working on Economics with Fable 5 by ekelsen on Tuesday September 08, 2026 @10:38AM (#49610956)
So show us the prompt
Re:Working on Economics with Fable 5 by estearum on Tuesday September 08, 2026 @11:20AM (#49611640)
As stated, it's work product. Not a toy "adversarial prompt," so I won't be posting it here.

And as stated, even Pangram's own marketing material claims at least a 0.4% false negative rate on their own contrived corpus. A one in 250 event is hardly even rare.

You seem quite defensive about Pangram, and it shows up in a few of your comments. Is there a reason for that? If it's extending to outright denial of Pangram's own stated error rates... sheesh... very weird indeed (unless they're paying you, obviously).

Re:Working on Economics with Fable 5 by ekelsen on Tuesday September 08, 2026 @01:56PM (#49614021)
What I said is that every time Pangram has said 100% AI, it is 100% something I don't want to read. I don't particularly care if it was written by humans or LLMs.

Every time I've asked for evidence from someone who claims something like you do there always seems to be a reason it can't be posted. So it makes one suspicious.

Is there no way you can modify your prompt to be shareable?

Re:Working on Economics with Fable 5 by estearum on Tuesday September 08, 2026 @02:09PM (#49614256)
> I never found a piece of text that pangram claims is 100% AI that to me seems like purely human written text.

That's another claim you made, which is quite silly. You've never found something that Pangram marks as 0% human and which you deem to be 100% human? Huge gulf.

And no, not really. It's copy from fundraising materials. Removing the identifiable portions of it would remove virtually everything, which is perhaps why Pangram didn't flag it?

Re:Working on Economics with Fable 5 by ekelsen on Tuesday September 08, 2026 @05:07PM (#49617061)
I said it wasn't something I wanted to read. Humans are capable of writing things I don't want to read. Which at this point is starting to include your comments.
Re:Working on Economics with Fable 5 by estearum on Tuesday September 08, 2026 @07:41PM (#49618731)
Did you not write the quoted text?
Re:Working on Economics with Fable 5 by ekelsen on Tuesday September 08, 2026 @10:59PM (#49620304)
I did and it's true, but I amended my statement at the top of this thread, which you've ignored.

What I wrote is true (it is a claim about my experience), so I'm not sure what you're attacking.

I amended my statement to avoid a discussion about the provenance which is not what I actually care about. What I care about is if I want to read it. Humans could theoretically write something in AI voice and if they did, I wouldn't want to read it.

I have not seen any examples of human text in LLM style that clearly predate 2023. If you can supply one, that would be incredible evidence.

If you're able to supply a prompt which generates text that fools pangram and is not like one sentence long or a list of names or something, but actual prose I want to read, then I will concede.

Re:Working on Economics with Fable 5 by estearum on Wednesday September 09, 2026 @08:59AM (#49625832)
Oh, well sure, but I don't care about your preferences for what you'd like to read. You can use a Ouija board to decide whether something is worth reading for all I care, just don't use it to accuse other people of plagiarism or dishonesty.

Duh.

If your statement was "my super special dowsing rods tell me what I want to read with 100% accuracy" I'd say "cool." But that wasn't your claim.

> If you're able to supply a prompt which generates text that fools pangram and is not like one sentence long or a list of names or something, but actual prose I want to read, then I will concede.

This is an even more conservative claim than the one you just made! Sheesh.

Have a good day.

Re:Working on Economics with Fable 5 by dwaltrip on Tuesday September 08, 2026 @12:23AM (#49605589)
Pangram 4?
Re:Working on Economics with Fable 5 by estearum on Tuesday September 08, 2026 @08:23AM (#49609363)
Whatever is deployed on their website 13 hours ago
Re:Working on Economics with Fable 5 by jefftk on Monday September 07, 2026 @08:44PM (#49604429)
Yes? The author is very clear in the blog post about having used AI very heavily:

Up until now you have been reading the words of Anthropic’s Claude, in particular Fable 5. In part, this is because I simply could not have written this piece. ... Although it might seem strange I thought it wrong to impose too much upon the machine, changing its voice to emulate mine. Richard Sutton’s bitter lesson would likely advise as much. Instead it has its unique voice, grating to some perhaps but altogether fitting that it should be able to keep it, and I only gave it advice on what considerations during writing would bridge the gap between its understanding and that of a reader.

Re:Working on Economics with Fable 5 by ekelsen on Monday September 07, 2026 @08:48PM (#49604456)
I don't know if putting a disclaimer at the very end of the blog, that you also disclaim, should be considered "very clear".

"Here is, for those that have read this far, the acknowledgements that used to be in the formal paper but we decided to remove it, mostly because the paper has been reworked and rewritten so many times by us."