- False receipts are now easy to generate with chatgpt
- They are convincing and additional adjustments make them look very real
- Operai says that metadata and use policies will protect us
The generative AI seems to be fixing one of the most obvious problems with the images generated by AI: the text. And although that is good news for anyone fed up with seeing images containing absolute gays, it can also have unwanted consequences, it turns out that Chatgpt is now quite good for creating false receipts.
Falses are not completely convincing if you know what to look for: they are too clean, a bit too CGI and, sometimes, the sums are wrong (as in our previous example). But they are also easy to lose, and with a little violin you can make them look more realistic with folds and food stains.
For now, it seems that the most effective way to do so is to make Chatgpt recreate an existing receipt, which can then be changed to show the prices you want.
Operai says he is not too worried, but it seems to be a new effective tool for digital scams and uproar.
Why Operai agrees with falsifications, for now
You can use 4 to generate false receipts. There are too many real world verification flows that depend on “real images” as proof. That time is over. pic.twitter.com/9Fors1PWSBMarch 29, 2025
We communicated with Openai and a spokesman told us that he is attentive to trends such as receipts generated by AI. “We monitor the generations of images inside and outside our platform, we use internal tools to verify that they were created by our products and take measures when we identify violations of our use policies,” they told us.
“We are always learning from the use and feedback of the real world, and we will continue to refine our policies to balance creative freedom with the prevention of improper use. All images include standard C2PA metadata of the industry that indicate that they were generated by AI by OpenAI,” the Openai spokesman added.
However, Operai spokeswoman, Taya Christianson, went a little further, telling TechCrunch that there was a positive potential use for false receipts: “teach people about financial education” in non -fraudulent situations.
I can also imagine using it to illustrate articles, or to create accessories for a real or virtual staging. But false receipts can also be used for fraud, and the more convincing images, the more fraud they will probably facilitate.
The ‘metadata saves’ made by chatgpt’ are fine, but exporting an image without metadata is a single click job. I am not sure that these safeguards do the least to prevent false images from being used for fraud more than the policies of “being pleasant” of social networks have prevented people from being horrible on the Internet.
But I have to admit that it would be a lot of fun if Openai’s own employees began to use technology to inflate their expense claims.
You may also like
