Hey everyone,
I have noticed that some recipes on the internet make no sense and have my suspicions that they may be AI slop. The ratios are off, the cook time is unlikely, the illustration is AI made… But it’s hard. And it will likely be getting harder to identify AI generated recipes in the future.
It’s come to the point that I have a hard time trusting a recipe witten after the AI craze started (let’s call it 2025). I’m so suspicious of everything, my go-to “authenticity” check is to not bother with recipes that have recent publication dates. But this isn’t exactly fair nor fool proof.
Do you have tips on how to spot an AI generated recipe?
If you are using ublock Origin or a fork of it add these filters to your block lists to hide a lot of AI junk.
fyi it’s uBlock Origin!
Yea… I really should proofread my comments…
I typically stick to recipe websites with a good reputation rather than random blogs that come up in search results. But that can be tricky if you’re looking for a specific dish.
Might have to go back to the cookbook concept of yore.
So, either buy a physical, reputable cookbook and then try the recipes in there. Or just go to known-good recipe webpages and go through their recipes.
But yeah, just thinking “I’d like to make Venezuelan vuvuzela vegetables”, then typing that into a search engine and cooking whatever’s the first result, that won’t be possible anymore…
I’ve mentioned this elsewhere, but it’s a drum worth beating; if you live somewhere with a public library, there is a very good chance that they have a collection of cookbooks that are available to check out for free. Alternatively, no one has ever batted an eye at me making copies of the specific recipes I want to try right there at the library. I have to pay $0.10 a copy, but it’s worth it. Especially since I know how I cook, and this keeps the library’s books out of harm’s way. My local library has stuff ranging from the latest James Beard winners, to tried and true standbys like The Joy of Cooking, as well as stuff with a local focus (either ingredients, or historical).
Funny you mention the library; I put a couple of cookbooks on hold just a few minutes ago because of this community!
Yes, cookbooks are the way to go. SEO and AI slop have made finding recipes on the Internet a frustrating experience. I’ve slowly built up a collection of about five to seven cookbooks. Those books offer me plenty of choices. For example, just a few days ago, I made buttermilk biscuits and ended up having four different recipes to compare. I also find myself flipping through them to discover new recipes. It’s offline, peaceful, and I’m not bombarded with ads.
It’s kind of like music. I used to have a collection of cookbooks to do exactly that, just like I used to have a collection of music in various formats as technology changes. But it’s too limiting. It can never compare to a digital search among all the world’s recipes/music.
My approach was to buy a recipe manager. I generally search online to find something new that looks interesting. But then I import it into my recipe manager so I can use it without all the life story, the excessive ads, the really annoying screen redraws to show yet more ads, the popped to display additional ads or the inline partial recipes that are more ads in disguise
If I’m searching for it, I limit searches to before 2024.
If I’m browsing a recipe site, I ignore recipes posted after 2024.
Im sure I’m missing a lot of good stuff, but I don’t always know enough about cooking to know just by looking at the recipe that it doesn’t work.
I’d skip any with AI images for sure
If the page lacks a date (or has a recent one) then that’s a pretty big tell. Same if they’ve posted dozens of recipes per day. Basically the same as YouTube videos.
generally speaking, i think it’s good practice to find several recipes and compare and contrast them. you’ll find opinions and get a sense for what the writer’s priorities are (quick, fewer dishes, what they usually have in the pantry, etc) and can figure out which writer has similar priorities to you. or just synthesize a recipe from those sources. this does require some technical know-how, but i think this is a good skill to have.
I wonder if the first “AI generated” recipe was the birthday cake in Portal (actually written by humans, but recited by an AI who it’s insinuated made the recipe). This recipe includes broken glass, so you kinda know it’s not real.
At the end (not really a spoiler), they show a black forest cake, which is delicious. The recipe given obviously does not make a black forest cake, even setting aside the joke comments like the broken glass.
The problem I see is that sooner rather than later, the problems will be so small that most home cooks won’t catch the problem until it’s too late.
The bigger problem is that if experienced cooks can be taught it’s wrong (e.g. something like swapping baking soda for baking powder), why can’t the AI, assuming the AI’s goal is actually to help you. I feel like we are at a point with AI that if I ask for a recipe with very specific requirements, it should either be able to conjure up the recipe, or tell me why it can’t be done so I can change the parameters equivalent (e.g. you can’t ask for a steak to be vegan, that simply does not compute).
A simplified version of how LLM AI works is that it has “read” all the books in the world, and selects the most likely next word/sentence. This can look a lot like knowing, but there’s no reasoning or sythesis behind it, which is why it can’t be “taught” that baking powder leads to a more sour outcome than baking soda.
AI has no goals. The companies that program and release AI have the goals of keeping you engaged with the AI.
I’ve seen some with impossible cook times but I assume that’s just different ingredient expectations or something.
For example I want to look up a time and temperature to cook chicken breasts but the time is impossibly short. At that temperature it may take 2-3x the time. Are we already at the point of ai recipes with unnoticeable flaws or do “chicken breasts” mean something very different in different places? Or maybe it’s a flawed conversion from metric?
just different ingredient expectations or something
No, it’s fundamental to the way LLMs (don’t) work. Take 10 random pages from a cookbook. Look at the cook times. I’m guessing the “impossible” times you’ve noticed will be within the range of times from the random cookbook.
The LLM doesn’t actually know anything about cooking; it’s just mashing together something plausible based on 1000 previous cookbooks.
Maybe it’s pressure cooker times?
