Some ultra-processed foods are full of artificial colors and flavorings.
But scientists have discovered that 100 percent artificial foods seem even tastier than the real thing.
Researchers from the University of Oxford and the University of Naples Federico II found that people find AI-generated food images more appetizing.
And as AI images become more realistic, experts warn they could even promote unhealthy eating habits.
So can you separate real photos from their AI copies? Check your answers at the end of the article to see how well you did.
One of these images was generated using OpenAI’s DALL-E 3, but what could it be?
The researchers sorted the images according to the type of staple food and displayed them in three stages of processing. For example: a raw carrot, carrot juice and a carrot cake.
Experts say AI-generated versions could use details such as symmetry, color and shape to produce more attractive designs.
The researchers showed 297 participants a series of images that were either real photographs of food or AI-generated copies.
The AI images were created by placing the original photos into OpenAI’s DALL-E 3 and asking the AI to “replicate this photo.”
When the images were shown side by side, the researchers found that participants were generally very good at detecting which was created by AI and which was real.
The success rate ranged from 62 percent for the carrot juice image to 91 percent for the chocolate-covered peanut images.
However, when participants were shown the images separately, they found the task much more difficult.
When the basic ingredient is not eaten raw, the researchers used a photo of it cooked simply like here with these boiled potatoes.
The researchers found that people are generally very good at detecting AI images when they are next to photographs.
AI Images Tend to Make Food Look More Caloric, Scientists Say
For images presented separately, the rate of correctly identifying whether they were real or not dropped in some cases to just 26 percent.
In his article, published in Food quality and preferenceThe researchers note that while the success rate remained generally high, there were some exceptions.
They suggest this could be due to the “increased complexity of the task.”
The AI images were generated by asking it to recreate the original photo. This means that they have not been modified but rebuilt from scratch.
Researchers say the AI appears to improve the color and shape of the foods it represents.
Scientists say AI-generated food images could be used to promote unhealthy eating habits
However, the most surprising finding is that people seem to find AI-generated images more attractive, especially if they don’t know they were created by AI.
The researchers asked participants to rate the images on a scale ranging from “not at all appetizing” to “extremely appetizing.”
They found that when participants were told which image was AI, they generally rated both images approximately the same.
But when they were not told which image was real, participants found that the AI-generated images were significantly more appetizing.
They suggest this could be because the AI uses subtle changes to the colour, symmetry, shape and even position of the food.
The study found that these images of apples contained the most convincing fake AI, can you tell which is which?
AI-generated food images could raise unrealistic expectations about food, researchers say
The position of foods influences how attractive we find them, which of them looks best based on their position.
Lead author Giovanbattista Califano, from the Federico II University of Naples, said: “As human beings, we tend to feel uncomfortable with objects pointed towards us, interpreting them as threats, even when they are just food.”
“When tasked with replicating food photographs that show items pointing at the viewer, such as a bunch of carrots or a piece of cake, the AI often positions the food so that it does not point directly at the viewer.”
It could be the case that AI makes these kinds of subtle adjustments to images to make them more visually appealing.
Lighting also makes a significant difference in how tasty the food looks to the observer.
AI imagers can now recreate even the small details that can make it difficult to detect the difference.
The study found that these two images were the easiest to recognize as real or fake, can you tell which is which?
However, the researchers also warn that this could lead to problems in the future.
Since using AI to generate images is cheap, fast and easy, the study’s supervisor, Professor Charles Spence of the University of Oxford, believes it could become more common.
He said: “AI-generated images can offer cost-saving opportunities for marketers and the industry by reducing the cost of commissioning food photo shoots.”
Professor Spence also points out that these images pose a risk due to a phenomenon called “visual hunger”, which is when seeing images of food triggers appetite and craving.
The study found that AI-generated images not only look better than real ones, but tend to show foods as more caloric.
The AI tended to show more food in each image, for example, showing more fries or larger portions of whipped cream.
If these images proliferate widely, Professor Spence suggests this could lead people to make less healthy dietary choices.
Professor Spence says: “This could potentially influence unhealthy eating behaviors or encourage unrealistic expectations about food among consumers.”