Last April, in front of the U.S. House Committee on Energy and Commerce, Facebook CEO Mark Zuckerberg told Congress that the social media giant was “not collecting any information verbally on the microphone,” and “does not have contracts with anyone else who is.”
On Tuesday this week, Bloomberg reported that Facebook has indeed been paying outside contractors to transcribe clips of audio from users.
The work has rattled the contract employees, who are not told where the audio was recorded or how it was obtained — only to transcribe it, said the people, who requested anonymity for fear of losing their jobs. They’re hearing Facebook users’ conversations, sometimes with vulgar content, but do not know why Facebook needs them transcribed, the people said.
A Facebook company spokesperson told The Federalist, “Much like Apple and Google, we paused human review of audio more than a week ago,” confirming that the company had been transcribing users’ audio and chalking up the activity to common practice among tech companies. The company said nonpublic audio under human review would have only come from devices on which users had given permission to access microphones.
At the heart of the issue, and the key factor in determining how truthful Facebook has been in addressing conspiracies about its listening habits, is the specified purpose of audio review. In 2016, Facebook went on the record to state that it does “not use your phone’s microphone to inform ads or to change what you see in News Feed.”
During an exchange with Zuckerberg, Sen. Gary Peters asked, “Yes or no, does Facebook use audio obtained from mobile devices to enrich personal information about its users?”
Zuckerberg responded, “We do not. Senator. Let me be clear on this. You are talking about the conspiracy theory passed around that we listen to what is going on on your microphone and use that. We do not do that.”
Now, the company claims that while review of audio from users who allowed access did happen, the purpose of the audio review was strictly to improve the ability of artificial intelligence to transcribe messages, something the company never mentioned as a possible use of the audio in previous statements.
Similar privacy concerns have surfaced with the popularity of devices such as Google Home and Amazon’s Echo. In April, Bloomberg reported that Amazon employees and contractors manually review and transcribe clips of conversations users have with the digital assistant, “Alexa,” and that Amazon workers sometimes share certain audio snippets in internal company chat rooms.