How The ChatGPT Watermark Works And Why It Might Be Defeated

Posted by

OpenAI’s ChatGPT presented a way to automatically create content but plans to present a watermarking function to make it simple to discover are making some people worried. This is how ChatGPT watermarking works and why there may be a way to beat it.

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

Some online marketers like it because they’re finding new methods to utilize it to generate material briefs, details and intricate short articles.

Online publishers are afraid of the possibility of AI material flooding the search engine result, supplanting expert short articles composed by people.

Consequently, news of a watermarking function that unlocks detection of ChatGPT-authored material is also expected with anxiety and hope.

Cryptographic Watermark

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

It’s largely seen in photos and significantly in videos.

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

Scott Aaronson and ChatGPT Watermarking

A prominent computer scientist named Scott Aaronson was employed by OpenAI in June 2022 to work on AI Safety and Positioning.

AI Security is a research study field concerned with studying manner ins which AI may pose a damage to humans and producing methods to prevent that kind of unfavorable interruption.

The Distill clinical journal, featuring authors affiliated with OpenAI, defines AI Safety like this:

“The objective of long-term expert system (AI) security is to guarantee that advanced AI systems are reliably lined up with human values– that they reliably do things that individuals desire them to do.”

AI Positioning is the artificial intelligence field concerned with making sure that the AI is lined up with the designated goals.

A large language model (LLM) like ChatGPT can be used in a way that might go contrary to the goals of AI Alignment as defined by OpenAI, which is to produce AI that benefits humankind.

Appropriately, the factor for watermarking is to avoid the misuse of AI in a manner that harms humanity.

Aaronson explained the factor for watermarking ChatGPT output:

“This could be practical for preventing academic plagiarism, undoubtedly, however 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.

Material created by expert system is generated with a relatively predictable pattern of word option.

The words composed by humans and AI follow an analytical pattern.

Altering the pattern of the words used in generated content is a way to “watermark” the text to make it simple for a system to discover if it was the item of an AI text generator.

The trick that makes AI content watermarking undetectable is that the circulation of words still have a random appearance similar to normal AI produced text.

This is described as a pseudorandom distribution of words.

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

ChatGPT watermarking is not presently in usage. However Scott Aaronson at OpenAI is on record specifying that it is planned.

Right now ChatGPT remains in previews, which enables OpenAI to discover “misalignment” through real-world use.

Most likely watermarking may be introduced in a final version of ChatGPT or earlier than that.

Scott Aaronson blogged about how watermarking works:

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

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

Aaronson discussed further how ChatGPT watermarking works. However first, it is necessary to comprehend the principle of tokenization.

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

Tokenization modifications text into a structured type that can be used in machine learning.

The procedure of text generation is the device thinking which token follows based upon the previous token.

This is finished with a mathematical function that determines the probability of what the next token will be, what’s called a likelihood circulation.

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

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

Here is the technical description 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 have to do with 100,000 tokens in overall.

At its core, GPT is continuously generating a likelihood circulation over the next token to produce, conditional on the string of previous tokens.

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

As long as the temperature is nonzero, though, there will normally be some randomness in the option of the next token: you might run over and over with the exact same timely, and get a various conclusion (i.e., string of output tokens) each time.

So then to watermark, rather of choosing the next token arbitrarily, the idea will be to pick it pseudorandomly, using a cryptographic pseudorandom function, whose secret is known only to OpenAI.”

The watermark looks totally natural to those checking out the text due to the fact that the choice of words is imitating the randomness of all the other words.

But that randomness consists of a bias that can only be identified by someone with the secret to decipher it.

This is the technical description:

“To illustrate, in the diplomatic immunity that GPT had a bunch of possible tokens that it judged similarly likely, you could simply choose whichever token taken full advantage of g. The option would look evenly random to somebody who didn’t know the secret, but somebody who did understand the key could later 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 could keep a record of every output it creates and utilize that for detection.

Scott Aaronson confirms that OpenAI might do that but that doing so positions a privacy problem. The possible exception is for police circumstance, which he didn’t elaborate on.

How to Identify ChatGPT or GPT Watermarking

Something fascinating that seems to not be popular yet is that Scott Aaronson kept in mind that there is a method to beat the watermarking.

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

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

For example, if you utilized another AI to paraphrase GPT’s output– well okay, we’re not going to have the ability to identify that.”

It seems like the watermarking can be beat, at least in from November when the above declarations were made.

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


Check out Scott Aaronson’s article here.

Featured image by SMM Panel/RealPeopleStudio