Article image

AI-generated food images are more enticing than real images

Researchers have made a compelling discovery: when unaware of the images’ origins, consumers show a marked preference for these AI-generated visuals over real food photos.

Detailed in the journal Food Quality and Preference, this revelation urges us to reassess the influence of visual appeal on our food choices.

Secret ingredients of AI food imagery

The study shows AI-generated food images captivate by optimizing visual features like symmetry, shape, glossiness, and light and color interplay. These elements are pivotal in making food imagery more appealing.

Giovanbattista Califano, the lead author and a researcher from the University of Naples Federico II’s Department of Agricultural Sciences, sheds light on an interesting nuance: the positioning of food items.

“As humans, we naturally perceive objects pointing towards us as potential threats, a reaction that extends even to inanimate objects like food,” Califano said.

The AI cleverly adjusts the positioning of items, such as a pointing bunch of carrots or a slice of cake. By doing so, it makes them appear less direct and, thus, more inviting.

The taste test: AI vs. reality

In their study, the researchers engaged 297 participants. They asked them to rate various food images on their appeal, ranging from natural items like apples to ultra-processed foods like chocolate milkshakes.

Interestingly, the twist came when some participants were informed about the origin of these images, while others were not.

Subsequently, the findings were revealing. Without knowing the images were AI-generated, participants found them significantly more appetizing than their real counterparts.

Consequences of visual hunger

Professor Charles Spence, a co-author and study supervisor from the University of Oxford’s Department of Experimental Psychology, points out a dual-edged sword.

On one hand, AI-generated food visuals present a cost-effective alternative for marketers, potentially reducing the need for expensive photoshoots.

On the other, they risk amplifying “visual hunger” — the phenomenon where merely viewing food can trigger cravings and appetite. This could, in turn, influence unhealthy eating habits and set unrealistic food expectations among consumers.

Moreover, the research underscores a tendency of AI images to depict food as more energy-dense. This often involves exaggerating the quantities of food, such as adding extra fries or more whipped cream to desserts.

Given our natural draw to energy-rich foods, this raises concerns over promoting unhealthy eating habits due to cue-induced cravings.

Ai images and the impact on food perceptions

The implications extend beyond individual health, touching on broader environmental concerns. As the world gravitates towards more sustainable eating habits, including the acceptance of ‘ugly’ fruits and vegetables, a new challenge emerges.

The prevalence of idealized, AI-enhanced food images could set unrealistic standards for natural food appearance, potentially undermining sustainability efforts.

This study highlights the influence of visual aesthetics on our food preferences. Additionally, it prompts a critical examination of how technology, specifically AI, shapes our perceptions and behaviors.

As we navigate the intersection of nutrition, technology, and sustainability, these findings provide valuable insights. They are relevant for marketers, policymakers, and consumers, urging a balanced approach to using and consuming AI-generated food imagery.

The full study was published in the journal Food Quality and Preference.


Like what you read? Subscribe to our newsletter for engaging articles, exclusive content, and the latest updates. 

Check us out on EarthSnap, a free app brought to you by Eric Ralls and


News coming your way
The biggest news about our planet delivered to you each day