How The ChatGPT Watermark Works And Why It Could Be Defeated

Posted by

OpenAI’s ChatGPT introduced a method to automatically develop material but prepares to introduce a watermarking function to make it easy to spot are making some people worried. This is how ChatGPT watermarking works and why there may be a way to defeat it.

ChatGPT is an extraordinary tool that online publishers, affiliates and SEOs all at once enjoy and fear.

Some online marketers love it due to the fact that they’re discovering new methods to utilize it to produce material briefs, describes and intricate articles.

Online publishers are afraid of the prospect of AI material flooding the search results, supplanting specialist posts composed by humans.

As a result, news of a watermarking feature that unlocks detection of ChatGPT-authored content is also anticipated with anxiety and hope.

Cryptographic Watermark

A watermark is a semi-transparent mark (a logo or text) that is ingrained onto an image. The watermark signals who is the initial author of the work.

It’s mostly seen in photographs and progressively in videos.

Watermarking text in ChatGPT includes cryptography in the form of embedding a pattern of words, letters and punctiation in the type of a secret code.

Scott Aaronson and ChatGPT Watermarking

An influential computer scientist named Scott Aaronson was worked with by OpenAI in June 2022 to deal with AI Safety and Alignment.

AI Security is a research study field interested in studying manner ins which AI might posture a harm to people and developing methods to prevent that kind of negative disturbance.

The Distill scientific journal, including authors connected with OpenAI, specifies AI Safety like this:

“The goal of long-term artificial intelligence (AI) safety is to make sure that innovative AI systems are dependably aligned with human worths– that they dependably do things that people want them to do.”

AI Alignment is the expert system field concerned with making certain that the AI is lined up with the desired goals.

A big language design (LLM) like ChatGPT can be utilized in a manner that might go contrary to the objectives of AI Alignment as defined by OpenAI, which is to develop AI that benefits humanity.

Appropriately, the reason for watermarking is to avoid the misuse of AI in a manner that damages humankind.

Aaronson described the reason for watermarking ChatGPT output:

“This could be handy for avoiding academic plagiarism, undoubtedly, but likewise, for instance, mass generation of propaganda …”

How Does ChatGPT Watermarking Work?

ChatGPT watermarking is a system that embeds a statistical pattern, a code, into the choices of words and even punctuation marks.

Content produced by expert system is produced with a fairly predictable pattern of word choice.

The words written by people and AI follow a statistical pattern.

Changing the pattern of the words used in produced material is a way to “watermark” the text to make it easy for a system to spot if it was the product of an AI text generator.

The technique that makes AI content watermarking undetected is that the distribution of words still have a random look similar to normal AI created text.

This is described as a pseudorandom circulation of words.

Pseudorandomness is a statistically random series of words or numbers that are not actually random.

ChatGPT watermarking is not currently in use. Nevertheless Scott Aaronson at OpenAI is on record mentioning that it is prepared.

Right now ChatGPT remains in sneak peeks, which permits OpenAI to find “misalignment” through real-world use.

Most likely watermarking might be presented in a final variation of ChatGPT or quicker than that.

Scott Aaronson discussed how watermarking works:

“My primary project so far has been a tool for statistically watermarking the outputs of a text model like GPT.

Basically, whenever GPT produces some long text, we desire there to be an otherwise undetectable secret signal in its options of words, which you can use to prove later on that, yes, this originated from GPT.”

Aaronson described further how ChatGPT watermarking works. But first, it is essential to understand the idea of tokenization.

Tokenization is a step that occurs in natural language processing where the maker takes the words in a file and breaks them down into semantic units like words and sentences.

Tokenization changes text into a structured kind that can be utilized in machine learning.

The process of text generation is the maker thinking which token comes next based on the previous token.

This is made with a mathematical function that figures out the possibility of what the next token will be, what’s called a possibility circulation.

What word is next is predicted however it’s random.

The watermarking itself is what Aaron refers to as pseudorandom, because there’s a mathematical reason for a particular word or punctuation mark to be there however it is still statistically random.

Here is the technical explanation of GPT watermarking:

“For GPT, every input and output is a string of tokens, which could be words but likewise punctuation marks, parts of words, or more– there are about 100,000 tokens in overall.

At its core, GPT is constantly generating a possibility circulation over the next token to create, conditional on the string of previous tokens.

After the neural net creates the circulation, the OpenAI server then really samples a token according to that distribution– or some modified version of the distribution, depending upon a criterion called ‘temperature.’

As long as the temperature level is nonzero, though, there will typically be some randomness in the choice of the next token: you could run over and over with the very same timely, and get a different conclusion (i.e., string of output tokens) each time.

So then to watermark, instead of picking the next token randomly, the idea will be to select it pseudorandomly, using a cryptographic pseudorandom function, whose key is understood just to OpenAI.”

The watermark looks entirely natural to those checking out the text since the choice of words is imitating the randomness of all the other words.

However that randomness consists of a bias that can just be discovered by somebody with the secret to decode it.

This is the technical description:

“To highlight, in the special case that GPT had a bunch of possible tokens that it evaluated similarly probable, you could just choose whichever token made the most of g. The option would look evenly random to someone who didn’t know the key, however somebody who did understand the key could later on sum g over all n-grams and see that it was anomalously large.”

Watermarking is a Privacy-first Service

I’ve seen discussions on social networks where some individuals suggested that OpenAI might keep a record of every output it produces and use that for detection.

Scott Aaronson verifies that OpenAI could do that but that doing so poses a personal privacy problem. The possible exception is for law enforcement situation, which he didn’t elaborate on.

How to Discover ChatGPT or GPT Watermarking

Something fascinating that appears to not be well known yet is that Scott Aaronson kept in mind that there is a method to defeat the watermarking.

He didn’t say it’s possible to beat the watermarking, he stated that it can be beat.

“Now, this can all be defeated with adequate effort.

For instance, if you used another AI to paraphrase GPT’s output– well all right, we’re not going to be able to identify that.”

It looks like the watermarking can be defeated, a minimum of in from November when the above statements were made.

There is no indicator that the watermarking is presently in use. But when it does enter usage, it might be unidentified if this loophole was closed.

Citation

Read Scott Aaronson’s blog post here.

Included image by Best SMM Panel/RealPeopleStudio