Viral Post Shows How AI Image Of Cracked Eggs Was Allegedly Used For Instamart Refund

As more and more people around the world adopt AI tools for various purposes, instances of misuse also seem to be rising. Social media is filled with examples of how AI-generated images and videos are being used to mislead and manipulate. Recently, a post on X (formerly Twitter) highlighted a specific way in which AI images can not only diminish trust but also cause real financial problems for companies. The now-viral post claimed that an Instamart user used an AI-generated image to get a refund for an order.

As per the post, the user had ordered a large box of eggs via the quick commerce platform. When the package arrived, one of the eggs was cracked. Instead of flagging this issue, the user allegedly decided to magnify it. They used an AI tool to make it seem like more eggs were cracked. The post explains, "Instead of just reporting it, they opened Gemini Nano and literally typed: 'Apply more cracks.' In a few seconds, AI turned that tray into 20+ cracked eggs - flawless, realistic, impossible to distinguish."

Also Read: 'Indian Mom - 1, Gemini - 0': Gemini Fails To Guess What's Inside Containers In Indian Kitchen

The post includes photos of the user's chat with the AI tool as well as their chat with Instamart's customer service. In the latter, we can see that they used the AI-modified image to claim a full refund. Customer support was fooled by the 'proof' provided and said that a refund of Rs 245 would be processed within 2 hours.

The viral post continues, "Just pause and think about that. Our refund systems were built for a world where photos were trustworthy. But now they're up against 2025-level AI - and they're getting absolutely destroyed. If even 1% of people start doing this, quick-commerce unit economics won't just suffer - they'll implode. AI isn't the villain here. The real problem is verification systems stuck in the past. Welcome to the era of AI vs AI."

Someone ordered eggs on Instamart and only one came cracked.

Instead of just reporting it, they opened Gemini Nano and literally typed:
“apply more cracks.”
In a few seconds, AI turned that tray into 20+ cracked eggs — flawless, realistic, impossible to distinguish.

Support… pic.twitter.com/PnkNuG2Qt3

— kapilansh (@kapilansh_twt) November 24, 2025

Disclaimer: NDTV does not vouch for the claims in the post by the X user.

In the comments, people had a lot to say about the viral post. Some suggested that companies adopt AI-detection tools soon. Others expressed concern about what such incidents reflect about our society and our future. Read some of the reactions below:

AI generated photos have a digital signature that can be tracked and identified and I am not talking about logo.
My guess is that at some point Gemini will also have an enterprise API to identify and weed out AI generated images.
Though what I am not clear is if different tech…

— Kushagra Agarwal (@kushagra_26) November 24, 2025

Open box delivery with go pro should be common for all delivery folks now @Swiggy @zomatocare @ZeptoNow @letsblinkit

— Dick Bonas (@cs_walla) November 24, 2025

Google generated AI images uses SynthID and you can upload to Gemini to check whether it's AI generated or not.
Every e-commerce and q-commerce platform will have to integrate it sooner.

— Meet (@MeetSiddhapura) November 24, 2025

When proof becomes editable, trust becomes a vulnerability.

One fake refund doesn't matter.
Millions will break the business.

This isn't about cracked eggs.
It's about cracked systems.

— City of Stars (@IamFatima5) November 24, 2025

If operations don't adapt, fraud will scale faster than revenue.

— Akshay Purohit (@iamakkshayy) November 24, 2025

just a low trust society , trying to scam a company bcz they feel they are entitled to refunds .

— Crosshair (@crosshair_io) November 24, 2025

Most AI-generated photos may soon include hidden signatures. Many companies that create AI images and videos are expected to embed such markers, making it harder for the human eye to detect whether an image is real or not. But when you create an image using Gemini, it leaves a… pic.twitter.com/UGEC4gtf5Z

— Raghu Sahay (@raghusahay) November 24, 2025

Apps like these should only allow clicking pictures directly through their camera and stop allowing upload of images from gallery.

— Pritish Poswal (@pritishposwal) November 24, 2025

Also Read: Indian Woman Asks ChatGPT To Pick A Watermelon. What Happens Next Divides The Internet

Another similar incident took social media by storm some days ago. A Zomato customer attempted to fraudulently claim a refund of Rs 1,820 from Mumbai-based bakery Dessert Therapy by submitting an AI-generated image of a damaged premium cake. The bakery exposed the scam on Instagram, identifying clear AI errors such as unnatural textures and a misspelt "Appy Birthda" tag on the cake. The brand highlighted this as a blatant misuse of technology. The post received a lot of attention online, and people condemned the customer's unethical breach of trust. Read the full story here.