Think back to the first wave of internet amusement when AI tools began dabbling in the culinary sphere. People enthusiastically turned to these sophisticated systems for kitchen guidance, only to be met with absurd and sometimes alarming suggestions—such as the infamous advice to use glue as an adhesive for keeping cheese on pizza. That early incident, equal parts humorous and horrifying, became emblematic of how overly trusting we can be when it comes to machine-generated recommendations. Now fast-forward to the present moment: remarkably, many people are once again relying on the very same kind of technology, this time to orchestrate one of the most important and tradition-laden meals of the year—Thanksgiving dinner. According to a recent report by Bloomberg, the trend has grown so widespread that it’s measurably reducing the audience engagement for seasoned recipe creators, writers, and food bloggers who ordinarily enjoy a huge spike in readership during the holiday season.
The underlying issue has not changed since earlier controversies like the time Google’s algorithmic summaries famously advised users to consume a rock each day. The culprit in both cases is the same flawed mechanism—AI Overviews in Search. This tool is designed to synthesize content from across the web and present it in an easily digestible panel, supposedly highlighting only the most relevant information. However, this convenience comes at a considerable cost: it eliminates the need for readers to visit the original websites, effectively discouraging them from scrolling through full posts, including the long but often insightful personal essays that typically introduce each recipe. While those essays are frequently mocked for their length, they often contain important contextual details, testing notes, and safety tips that give meaning and accuracy to the recipe itself.
Two major problems arise from this situation. The first involves the creators themselves—the dedicated cooks, chefs, and writers who pour time, energy, and accumulated expertise into crafting and refining every recipe they share online. Their livelihoods depend on traffic, ad revenue, and the exposure that loyal readers bring. With AI Overviews siphoning off potential visitors, many of these professionals are witnessing catastrophic declines in web traffic. According to the Bloomberg interviews, numerous recipe authors have reported drops ranging from forty to eighty percent compared to previous Thanksgiving seasons. These figures align closely with wider industry data suggesting that websites across the board have suffered up to an eighty percent reduction in click-through rates since AI Overviews began dominating search results. In effect, the technology that was promoted as a tool for efficiency now threatens to undermine an entire ecosystem of digital creators.
The second, and arguably more alarming, issue lies with the users—the home cooks who trust these algorithm-generated summaries to guide them through meal preparation. Artificial intelligence, sophisticated though it may appear, does not truly comprehend the information it processes. It lacks any form of judgment, experience, or sensory understanding. It can only regurgitate patterns of text that statistically appear relevant, without discernment of accuracy or context. In an activity as precise as cooking, where minor errors in temperature, timing, or proportions can completely ruin a dish, such blind reproduction can have disastrous results. As Bloomberg describes, one baker’s popular Christmas cake recipe originally instructed readers to bake the cake at 160°C (320°F) for ninety minutes. The AI-summarized variant, however, erroneously recommended tripling or quadrupling the baking time—three to four hours in total. Anyone with even a modest grasp of baking can predict the consequence: a blackened, desiccated cake instead of a moist festive masterpiece.
The phenomenon has evolved into something of a miniature industry. Across countless social media platforms, users in search of inspiration may stumble onto visually appealing pages that imitate the tone and structure of authentic food content. Yet a closer look often reveals subtle oddities—cooking steps that seem off, ingredient combinations that don’t quite make sense, or cooking times that defy logic. The best possible outcome in these cases is relatively harmless: a dish that turns out lackluster or underwhelming, perhaps edible but devoid of flavor and finesse. The worst-case scenario, however, can be truly dangerous. Imagine following what seems to be a trustworthy AI-generated recipe only to discover that it recommends placing a piece of fish—wrapped tightly in tinfoil—into a microwave set to high power. The result could easily escalate from culinary disappointment to a full-blown household fire.
In light of these technological misadventures, perhaps a bit of old-fashioned wisdom is in order. As the holiday season approaches, it might be prudent to dust off those timeworn cookbooks resting on your kitchen shelf—pages stained with years of experience and marked with handwritten notes from parents or grandparents. Unlike algorithmic guesswork, those tangible volumes represent accumulated human understanding and genuine care. Trusting them may not only preserve your dinner from disaster but also honor the shared traditions that no artificial intelligence has yet learned to replicate.
Sourse: https://gizmodo.com/a-new-way-to-ruin-thanksgiving-making-ai-slop-recipes-2000691622
