The AI-powered cooking recipes that have started to flood search engines mix recipes from multiple pages and different chefs. The result, although coherent at the language level, is far from good when it comes time to put it into practice.
The so-called Generative Artificial Intelligence has recently found a new area of expansion, far from employment and social networks: the kitchen at home.
The proliferation, on search engines and digital platforms, of cooking recipes created by AI systems is changing the habits of millions of users, and raises a problem which, until now, had gone unnoticed by a large part of the public.
Since technology giants like Google have activated new search modes or suggestions made with AI, users have been receiving complete recipes no need to visit specialized websites.
These proposals do not come from a single source: they are built from the combination of excerpts from different human creatorsautomatically synthesized.
The result tends to be a impoverished version of the original recipesreduced to basic lists of ingredients and generic steps. The sentences seem to make sense, the ingredients and quantities match well-known recipes.
Mas a IA does not test or cook his recipesand his “knowledge” of gastronomy is limited to identify textual patternswhich prevents her from evaluating essential technical nuances so that a dish works in practice.
In many cases, the “frankenstein recipes” result in a tasteless, tasteless, meaningless porridge. In the worst case scenario, they can pose a health risk. It is “AI slop”, the mud of AI, served on your plate.
Automatic recipes, real mistakes
This limitation has generated obvious errors. Some systems even confuse forum comments with reliable sources, producing wrong and, in some cases, potentially dangerous recommendations.
The problem is not specific: it is a structural problem, notes the , because AI has no culinary criteria or direct experience.
For those gastronomic content creatorsthe impact goes beyond quality. Over the years, built reputations based on repeated testing, detailed explanations and technical knowledge.
This distinctive value disappears when your work is diluted in a synthetic recipe that does not reflect your real contribution.
The economic model is also affected. Most of these projects rely on digital advertising and web traffic. If the user obtains the information directly from the search engine, they stop visiting the original page — even if the its content has been used to train the system or to generate the automatic response.
The legal framework offers little protection. Recipes, as instructions, are not covered by copyrightwhich allows its reuse without direct attribution. Although the concrete wording may be protected, the logic of the dish is exposed to being replicated without compensation.
Nonetheless, distrust in relation to content grows generated by AI. As users interact more with these responses, the perception of lack of authenticity and real testing.
The kitchen has become a new front in the battle of automation against human experience. But having an AI come up with a cooking recipe seems, literally, a bad joke. Apparently, those responsible for the technological giants not only have no taste, they also lack sense.
