I asked ChatGPT to make a copycat Chick-fil-A recipe — but it left out a key ingredient
Winner, winner...not with this AI chicken dinner
Making dinner every night for a family of five is one of those things that has to be done even when I am exhausted and would rather order out. It doesn't help that my kids are picky eaters. One day they love what I make, the next day my middle schooler calls my cooking "mid."
And while I would love to order takeout more often, especially now that the GrubHub app is in Alexa+, it's just not in the budget. Plus, since ChatGPT has access to millions of recipes online, I figured it would be easy for it to help me recreate Chick-fil-A at home. Copycat recipes have existed on the internet for years. And AI is supposedly built to recognize patterns, combine information and deliver the “best” version of something instantly.
So naturally, I figured asking ChatGPT to recreate the perfect Chick-fil-A nuggets at home would be almost foolproof. Well, I was wrong.
I started by gathering the ingredients
A post shared by Amanda Caswell (@amandacaswell_ai)
A photo posted by on
What surprised me was that when ChatGPT gave me the recipe, it didn't cite any sources. I should have immediately taken this as a red flag. One Google search will tell you that there are plenty of Chick-fil-A copycat recipes, so it could have pulled from those. But the AI confidently delivered a recipe that looked accurate, and I foolishly took it at face value.
Instead, I bought the ingredients and thought about how pleased my family would be with such a fun and different dinner. And to my credit, I followed the recipe exactly as ChatGPT suggested. The instructions seemed correct.
The instructions were detailed. The measurements looked professional. It even explained why certain ingredients mattered, which made the whole thing feel even more trustworthy.
But after spending over an hour breading chicken, heating oil and trying to recreate one of the most recognizable fast-food sandwiches in America, I could just feel something was off.
ChatGPT was over confident and under delivered
Throughout the process, I used ChatGPT Voice as a guide. My hands were covered in breadcrumbs and I needed to know if I was doing everything correctly. The AI kept assuring me that yes, I was on the right track.
But, the problem came down to one tiny detail ChatGPT completely glossed over, especially since I had given ChatGPT a pretty specific prompt: “Give me the closest possible homemade Chick-fil-A nuggets recipe, including any secret techniques that make it taste authentic.”
Get instant access to breaking news, the hottest reviews, great deals and helpful tips.
After one bite, I realized that it completely missed the most important part of the entire recipe: the breading. It didn't give me any specific details about what type of breading to use, so I went with what I had. But after looking online at the copycat recipes with the best reviews, most offered very specific instructions about the breading. From Ritz crackers to sugar and even baking soda, none of those ingredients were given to me by ChatGPT.
Even though everything looked right, the first bite told me that AI was not the reliable chef I had hoped.
Using human creativity to fix AI's mistake
By the time I had finished frying all of the chicken, it was about 8:15 p.m. and there just wasn't time to start over with a new dinner. So, I covered the fried chicken in mozzarella and tomato sauce and baked it for a pathetic chicken parmesan. I was desperate at that point and luckily the family was hungry enough to eat it.
Probably the most important thing to remember is that ChatGPT stitches together the most commonly repeated phrases and repeated advice to present it like a fully formatted truth.
That works surprisingly well for many tasks. But cooking is sensory. Tiny details matter. Human experience matters. A person who has spent years perfecting fried chicken notices things AI simply can’t.
And, frankly, this bad experience made me rethink how I use AI. Sure, I still use ChatGPT constantly. But, this failed experiment remind me that AI is best as a starting point, not the final authority.
The takeaway
As AI becomes more integrated into our lives, it's times like these when I am forced to remember that it isn't an expert in anything. It's been trained on a lot of different things, but it doesn't have the very important human element that comes from experience, instinct and knowing when tiny details actually matter.
When has AI failed you? Share your worst experiences in the comments.
Follow Tom's Guide on Google News and add us as a preferred source to get our up-to-date news, analysis, and reviews in your feeds. Subscribe to Tom's Guide on YouTube and follow us on TikTok.
More from Tom’s Guide

Amanda Caswell is the AI Editor at Tom's Guide and one of today’s leading voices in AI and technology.
A celebrated contributor to various news outlets, her sharp insights and relatable storytelling have earned her a loyal readership. Amanda’s work has been recognized with prestigious honors, including outstanding contribution to media.
Known for her ability to bring clarity to even the most complex topics, Amanda seamlessly blends innovation and creativity, inspiring readers to embrace the power of AI and emerging technologies.
As a certified prompt engineer, she continues to push the boundaries of how humans and AI can work together.
Beyond her journalism career, Amanda is a long-distance runner and mom of three. She lives in New Jersey.
You must confirm your public display name before commenting
Please logout and then login again, you will then be prompted to enter your display name.
