DoorDash hopes to scale back verbally abusive and inappropriate interactions between consumers and delivery individuals with its recent AI-powered feature that robotically detects offensive language.

Dubbed “SafeChat+,” DoorDash is leveraging AI technology to review in-app conversations and determine if a customer or Dasher is being harassed. Depending on the scenario, there will probably be an choice to report the incident and either contact DoorDash’s support team for those who’re a customer or quickly cancel the order for those who’re a delivery person. If a driver is on the receiving end of the abuse, they’ll cancel a delivery without impacting their rankings. DoorDash may also send the user a warning to refrain from using inappropriate language.

The corporate says the AI analyzes over 1,400 messages a minute and covers “dozens” of languages, including English, French, Spanish, Portuguese, and Mandarin. Team members will investigate all incidents recognized by the AI.

The feature is an upgrade from SafeChat, where DoorDash’s Trust & Safety team manually screens chats for verbal abuse. The corporate tells TechCrunch that SafeChat+ is “the identical concept [as SafeChat] but backed by even higher, much more sophisticated technology. It will possibly understand subtle nuances and threats that don’t match any specific keywords.”

“We all know that verbal abuse or harassment represents the biggest style of safety incident on our platform. We imagine that introducing this feature could meaningfully reduce the general variety of incidents on our platform even further,” DoorDash adds.

DoorDash claims that greater than 99.99% of deliveries on its platforms are accomplished without safety-related incidents.

The platform also has “SafeDash,” an in-app toolkit that connects Dashers with ADT agents who can share the situation and other information with 911 services in an emergency.