Affiliate Disclosure


Readers like you help support AWG. If you buy through our links, we may get a commission at no cost to you. Read our affiliate policy.

Employees couldn’t handle it, and customers cried. Amazon has an idea for damaged goods

Share this article

[bsa_pro_ad_space id=4]

Amazon does not provide official statistics, however, according to unofficial data for the entire e-commerce industry, between 5 and 15 percent of all orders from online stores are refundable for various reasons. Most often, this reason is damage to the item, which affects 80 percent of all returns.

Losing money from the purchase is one thing, but the delivery of a damaged item also affects the entire customer-store relationship, which at best will be appeased, at some point returning to shopping, and at worst will leave a negative opinion and never make a purchase again.

In addition, not every damage can be avoided: the seller has no control over what happens to the product on its final path to the customer. But where control is, Amazon is implementing artificial intelligence that will check each product for damage before shipping to the customer.

Artificial intelligence will make sure that your order is not damaged. Amazon replaces people in the next field

The Wall Street Journal reported about another way to automate work in the mailing centres of the American e-commerce giant. According to the Daily, the process is already in the process of implementation and is intended to replace the laborious, time-consuming process of checking by human employees. Because every item, even in good condition, passes quality control before shipment. According to Amazon cited by the WSJ, on average, less than one item in a thousand shipped by Amazon is damaged.

Amazon’s artificial intelligence has been trained on a large collection of photos of both damaged and undamaged items so that it is able to capture the differences between the undesirable and desired condition of the product.

Artificial intelligence checks products during the collection and packaging process. Goods are selected for individual orders and placed in containers that pass through a special station, where the system checks the order to confirm that it contains a product that complies with the order. Amazon’s new AI will form part of this station and assess whether items in the order are damaged. If something is damaged, the container will be transferred to an employee who will further evaluate the order. If the system decides that the products are in accordance with the order and in the desired condition, the order will be transferred to packaging and sent to the customer.

Currently, artificial intelligence is already present in two fulfilment centres in the United States, but the giant also announces its implementation “in another ten in North America and Europe”.

Amazon is very eager to use developments in the field of AI. But…

This isn’t the first time Amazon has experimented with artificial intelligence. Last year, the company boasted a proprietary robot called Sparrow, which is able to detect, select and move individual objects, automating processes performed by people. The giant also does not ignore generative artificial intelligence, because it is working on an intelligent chatbot that is to replace the clumsy search on the platform.

At the same time, artificial intelligence is a double-edged sword for the company, as users passionately use it to write not very accurate opinions and earn benefits as part of the Amazon Vine program.

The Ultimate Managed Hosting Platform

Explore more on

Share this article

Harsh Singh

Harsh Singh

Hey there! I'm Harsh, a tech wizard and your digital guide. I geek out over the latest industry updates and make it my mission to help people navigate the online world safely. As the proud owner of a digital agency, I'm on a mission to empower businesses worldwide with top-notch digital services. Let's embark on this digital adventure together!

Alpha Web Guide Comment Policy

We welcome relevant and respectful comments. Off-topic or abusive comments will be removed. Please read our Comment Policy before commenting. Typo? Please let us know.