Meta has been mostly silent in the wake of revelations that it uses Ray-Ban owners’ naked videos to train AI, but if there’s one thing we know for certain, it’s that Meta CTO Andrew “Boz” Bosworth thinks that you chose for this to happen.
In one of Bosworth’s usual Instagram Live Q&A sessions from mid-March, he fielded a question that reads, “How private are Meta glasses? Who views the content I make?” His answer? Well, that’s up to you, silly.
“Every time content leaves the device, you have to have chosen for that to happen—either specifically or generally,” said Bosworth in his video response. “So, if you’re taking photos, those stay on your device. Now, if you’ve chosen to enable cloud storage, they go to the cloud. That’s up to you. If you’ve chosen to use AI, that goes to the server to get processed.”
And after you’ve “chosen” your path to capture content? Well, Bosworth expounds there, too:
“Every time you cross one of those thresholds, you, the consumer, are informed, ‘Hey, that means this data is going to the server, that means this can happen with it,’ that means you choose to engage with it or not,” Bosworth said. “But if all you want to do is grab the glasses and make content for yourself that you choose to share or not share, then you get to decide who views the content—that’s it.”
There are a lot of issues with that response, but probably the most obvious is the idea of consent. While Meta does “inform” users that its content is collected in certain scenarios, that stipulation is buried in its terms of service—it’s not what most would consider to be obvious. In fact, I think that’s pretty much the running definition of fine print.
Then there’s the issue of which content is allowed to be reviewed by Meta and which is not. Bosworth seems to suggest that if you want to keep content in your possession, all you have to do is choose not to share it, but according to Meta’s own rules, it’s not that simple. For example, any content that is captured by using the voice prompt “Hey Meta” is sent to the servers and is therefore subject to review. That’s not exactly intuitive, even if it is outlined in Meta’s dense terms of service.
Lastly, Bosworth doesn’t address the fact that there are reportedly humans reviewing content, not just robots. Those revelations, by the way, were reported weeks before this Q&A took place, so Bosworth certainly had an in to address them. That’s not even counting the fact that, in that report, human reviewers say they saw all kinds of sensitive content, including people naked, having sex, using the bathroom, and more. A portion of that content appears to have been recorded accidentally, too.
It’s a story as old as time, really; a company (Meta, in this case) buries something undesirable in its terms of service to cover its ass legally while hoping that no one notices it’s even there. And even when people do notice, little seems to budge. If you were wondering why a video from mid-March is still relevant, it’s because Meta has all but ignored the issue of privacy when it comes to smart glasses, despite calls from lawmakers, civil rights groups, and others to explain itself.
No matter, though, since, as Boz would frame it, you’re totally in control here. For sure. Yup, totally in control.
