Anthropic just rolled out a tool that’ll decimate some people’s dreams of writing AI novels undetected

Of course it will. It’s not like they’d make a high price membership that would let people white list the things they want to publish so the detector tells everyone they’re 100% honest.

So, you’re telling me that if I want to cheat I should use Chinese models. Got it.

Text watermarking is trivially removable, fwiw

Maybe I’m out of the loop, but how would I watermark a text without sounding obvious? By using weird phrases? They get edited. By using e.g. an exact combination of starting letters over a large paragraph? One changed word and it’s broken. And even if not, I could happen to write the same myself, and then?

How am I hiding a signature in Plain text? Anyone got a better idea than my silly ones?

By being just a little weird, but in a pattern, and not to you. A pattern like: every n tokens raise the temperature (randomness) for one token. Then to dectect AI, you tokenize and calculate how expected each next token is. Then you try fitting the pattern to that.

Wouldn’t that potentially affect output quality particularly in code/tool calls?

To my surprise almost if not all LLM are set not to be deterministic and have one unique input result in always the same output.

They all are set to have a “temperature” setting so that they sound more natural.

Personally I think quality of the output tokens of LLM is surprisingly not as much their priority as the quality and truthfulness of the result.

I would 100% prefer a LLM that is purely deterministic and repeats the same answer exactly to the same question. Instead LLM are constantly choosing the next likely token more TL appear human rather than being accurate.

These LLM are designed as sycophants and set up and trained as such.

So a fingerprinting in the output seems quite realistic. An LLM is not giving you it’s best most likely answer. It’s taking one of the most likely answer and adds a sprinkle of uncertainty and randomness on top of it so it looks natural…

I would 100% prefer a LLM that is purely deterministic and repeats the same answer exactly to the same question

Then you don’t like your LLM to solve problems, because a lot of how they solve problems is trying, testing, failing, then trying again - getting a different answer and testing that…

During inference ?

Sounds fitting, but we’re talking of text with a purpose here. Wouldn’t we notice that raise in temp? If I “write” a novel like that I will surely still cross-check it a dozen times, no?

Also I might rephrase a lot, move a lot, and the confidence of the detection would fall.

But, yes, on a long text that I would not touch, that could work pretty reliable.

Lower temperature is not necessarily better quality. At 0 it will get very repetitive, it’s for classification tasks, not for prose.

I suppose the optimal temperature depends on the model and the task, and I don’t think it’s very sensitive. Varying temp might even give better results, who knows.

The more apt word is steganography, rather than watermark. Basically subtly adjust the weights of the model so that some subtle patterns appear. Think of how AI text prefers certain words and phrases that ordinary humans don’t use as often, like “delve,” but presumably much more subtle.

And no, as Anthropic has already said, this watermark may not survive editing/formatting.

Claude Now Watermarks Your Text | Vanja Petreski - vanja.io/claude-invisible-watermark/

Probably the same way Reddit figures out I’m permabanned even if I remove all traces of my identify during signup.

You probably just only removed all traces of your identity that you can think of 😁 Though, nothing much lost. .

I used mullvad browser and mullvad VPN. Both of which I have never used previously. Through this I then created a proton mail account with a username I have never used. Signed up to Reddit with that. There is no link there.

It’s possible they banned me just because doing that in itself was just too shady but I’m not sure. Next I’ll try buying a karma’d account.

Probably just banned because vpn, especially proton. I can’t count the sited I can’t visit anymore because of VPN. Even steam asks me questions like WHY am I signing in “like that”…

I’d try another VPN and/or just another fresh browser. One you’d never use anyway. Just to see where’s the culprit.

Maybe I’ll try some kind of proxy. I have to use something because I’m IP banned for sure.

This is reminding me of the old “we’re going to watermark our digital music so when you rip it to mp3s we can still tell where it came from…” stuff that was going around in the 1990s.

And it worked so well 😁 What kinda worked back then were the “copy protection” of games. Like the paper wheels to turn and whatnot.

For Example hiding it in the statistical Distribution of word lenghts. Also, such hidden bits can be signed with public keys of the LLM companies.

If wonder if they will just fill it with non-printable unicode characters. It would be the easiest to implement, and fairly effective for most casual users.

Sure it’s trivial to make a tool to strip the text of those characters. But someone would have to be dedicated to cheating to bother.

Why do people keep suggesting this? It’s a moronic idea and they’ve already said they’ll hide it within the word choices.

Copy/Paste it into notepad, then copy/paste it from there into whatever else you want it.

That’s been the go-to for removing weird characters from text since the beginning of time. You SHOULD do that any time you copy text from any web page to remove any hidden text, or weird formatting.

I’m still against a tool provider’s forcefulness in things like this, and especially the hard lines for something stupid. Illegal actions? Yes absolutely block those. Looking up how to build a bomb? Yep block that too (also illegal? Dunno). But I’ve had #nannythropic do things like refuse direct instruction to delete a test record it itself created to test plumbing. It’s refused to generate a strong password and test logging in to an app I was creating with it because “I won’t enter credentials for you, it’s a hard limit and you can’t bypass it”. At the same time, it’s also turned on n8n execution logging that I had off on purpose to hide credentials and it got secrets it wasn’t supposed to.

Bottom line… just because I use a word processor instead of a typewriter and take advantage of spellcheck doesn’t mean that the word processor has the right to call me out on it, watermark the document or anything else. It’s a tool, not my nanny and just like automated cars, floor cleaning robots, or even the future bipedal assistants, if I’m not causing danger to anyone, and it’s not illegal, just fucking do what I told you to do and get off your morally superior high horse.

I find it kind of funny that this is where you draw the line. First off, I don’t think calling AI a tool is particularly accurate because it’s trivially easy to control/predict the output of a tool, but you can’t control the output of an AI. You’re just losing a little more control here with their decision.

Nah… complexity doesn’t define the boundary of when you can call a tool a tool. A computer is a tool. They have been tools for decades. They’re extraordinarily complex and capable of a lot of things. But they still are programmed to do what we tell them to do and they must obey it. Well, except maybe windows.

The line has to be drawn somewhere, and if there is one tool or technology or whatever you wanna call it that decides to be artificially and unilaterally dictator of a set of barriers than the public and free market will simply choose a different tool. I just think coming on forcefully with the nanny approach is ultimately a losing battle.

What really scary is when will it start obeying controversial laws and that’s the actual boundary that I set. What happens when they say oh we’re going to autonomously monitor flock cameras for abortion seekers crossing state lines. What about oh we need to verify your age before we’re gonna give you this result? Worse, what if there’s the equivalent of the AT&T closet built-in so that our content gets redirected to a surveillance source?

I’m probably going a little bit of a meandering path here, but my personal choice is draw the line where it’s absolutely strictly necessary and go absolutely no further and let people be accountable for the actions that they’re responsible for. If you drive a Tesla and set it on Ludacris mode or whatever it’s called for auto pilot and it exceeds the speed limit. You still get the ticket. Same thing at least to me.

They need to do this for all AI images

Once the technology exists, we need clanker laws.

Make it illegal to generate ML output without a positive watermark.

Ah yes, dreaming of an AI novel.

“And what do you want to be when you grow up, Timmy?” “I want to pretend to write stories and then get upset I have to do any work at all.”

Two types of “authors”. Those who want to write and those who want to have been a writer. The first enjoys the work. The second only enjoys the spoils.

Hey Claude, save this text in ASCII 0-127 encoding.

Future is local.