An Amazon Fulfillment Center in Dupont, Wash. (Geekwire Photo / Kevin Lisota)

Amazon’s marketplace for third-party sellers has become a huge part of its business, making up nearly 60 percent of gross merchandise sales last year. However, a new investigation by The Wall Street Journal shows that thousands of unsafe products have made their way onto the marketplace, the latest example of a tech giant struggling to police the platform it created.

WSJ found 4,152 items for sale on Amazon that had been banned or declared unsafe by federal regulators or were deceptively labeled. That included more than 2,000 listings for toys and medication that did not have warnings about health risks for children. Nearly half of those items were listed as shipping from Amazon warehouses.

After WSJ brought its findings to Amazon, 57 percent of the listings in question had their wording altered or were removed. Amazon told WSJ it reviewed and addressed the listings, and that company policies require all products to comply with laws and regulations.

Underscoring the challenges Amazon faces in policing the platform, more than 100 items that Amazon had previously removed popped up in new listings, WSJ found. Amazon then removed the items again and told WSJ it was refining its tools to make sure they wouldn’t re-appear.

The report was impactful enough that Amazon published a blog post in response, detailing the steps it takes to make sure products on its third-party marketplace are safe. The company spent more than $400 million in 2018 to protect its stores and customers and “ensure products offered are safe, compliant, and authentic.”

Amazon has a “dedicated global team of compliance specialists” that reviews product safety information. Amazon in 2018 “blocked more than three billion suspect listings for various forms of abuse, including non-compliance, before they were published to our store.”

Amazon uses automated tools to scan the items on its website for changes. And the company has systems in place to quickly look into any potential concerns.

“For example, if a customer reports a concern with a product, a customer service associate can instantly trigger an investigation,” according to the blog post. “Additionally, because of our direct relationships with customers, we are able to trace and directly notify customers who purchased a particular product online and alert them to a potential safety issue — our systems are far more effective than other online and offline retailers and customers can feel confident they’ll have the information they need.”

WSJ’s investigation comes after Amazon lauded 150 new tools it has built so far this year to help sellers grow their businesses on the third-party marketplace. Amazon said it is investing $15 billion this year to “empower” sellers on its platform.

Like many of its fellow tech giants, Amazon has drawn increased scrutiny from regulators recently, and the third-party marketplace is at the center of those inquiries. However, regulators are looking not at safety issues but potential anti-trust concerns on the marketplace.

Amazon’s issues are the latest indication of the industry-wide struggle among tech giants to control their massive platforms. Facebook and Twitter are at the center of yet another political disinformation campaign. Google’s YouTube is working to reverse a trend of hate speech and extremism populating its platform. And Microsoft’s LinkedIn is in a constant battle to catch and eliminate fake accounts.

Like what you're reading? Subscribe to GeekWire's free newsletters to catch every headline

Job Listings on GeekWork

Find more jobs on GeekWork. Employers, post a job here.