Facebook’s text virtual assistant, Facebook M – which relied on humans to train the artificial intelligence system – will lose the human help component of the system.
According to a report in The Verge, two and a half years after Facebook M was rolled out, the human-enabled portion – which users can access via a bot in Facebook Messenger – is being discontinued on Jan. 19. The company reported that contractors who worked on Facebook M will receive positions in other areas of the social media giant’s business.
Facebook isn’t killing off the feature completely, noted the report: Facebook M will still be available to make fully automated suggestions for things like making payments, planning outings and sending stickers to other users via Messenger.
“We launched this project to learn what people needed and expected of an assistant, and we learned a lot,” the company said in a statement. “We’re taking these useful insights to power other AI projects at Facebook. We continue to be very pleased with the performance of M suggestions in Messenger, powered by our learnings from this experiment.”
Facebook M isn’t the only area in which the social media giant has been using AI. In April, the company announced a new feature that will make recommendations to Facebook users during private conversations. Through the use of artificial intelligence, the feature will be able to detect language for recommending things, like getting an Uber, or sending a reminder to pay back a friend. Facebook claims this new feature is not meant to be invasive, but is intended to help improve the private chatting experience.
David Marcus, Facebook’s head of messaging products, commented to Recode about this new development: “The system learns from the things you like to do and you don’t like to do. If you don’t use the things that are being suggested for a specific use case, gradually those things will go away.” The Facebook team also shared that all messaging content will remain private and not used for targeted advertisement.