Complete strangers could be identified with even their addresses tracked down just by looking at them with smart glasses.
Two students from Harvard University built a programme allowing them to feed video from Meta’s high-tech Ray Bans for AI analysis.
As people walked by them, the glasses streamed visual information which was then monitored by an LLM (large language model AI).
They were able to ID dozens of people using reverse face search, and from there they were able to find addresses, phone numbers, and even people’s relatives automatically.
If Facebook founder Mark Zuckerberg is right about smart glasses becoming the next smartphones, the thought of ever disappearing into the crowd could soon be a quaint idea of the past.
Harvard student AnhPhu Nguyen, who built an app to use the glasses in this way, said in a demo video on X: ‘We stream the video from the glasses straight to Instagram and have a computer program monitor the stream.
‘We use AI to detect when we’re looking at someone’s face, then we scour the internet to find more pictures of that person.
‘Finally, we use data sources like online articles and voter registration databases to figure out their name, phone number, home address and relatives’ names.’
Fellow student Caine Ardayfio, made the app alongside him, added: ‘Using our glasses, we were able to identify dozens of people, including Harvard students, without them ever knowing.’
They published their project called I-XRAY online, though they did not make the technology available as their intention was only to show it was possible – and warn people of the potential problems.
How to stop this happening to you
This technology relies on publicly available data sources such as PimEyes and FastPeopleSearch.
To stop ordinary people, or scammers, finding you on this sites you can request they remove your data.
‘Face search engines’ let people upload a photo of a face and find photos elsewhere on the internet where that person appears.
You can use the links above to request removal of your face from the database.
AnhPhu Nguyen & Caine Ardayfio also collected the opt-out links to some of the major people search engines below:
They also advised people set up two factor authentication on their sensitive apps, such as banking apps, to make them more difficult to access even if someone can find your data.
Although removing yourself from public databases could protect you from amateur efforts such as this one using smart glasses, it will not be enough to shield you from sight of organisations with more extensive powers of search such as police or government, however.
Such abilities would be a dream for scammers, who could identify personal information about strangers and use it to build a rapport, such as by claiming they had seen them somewhere before, which could bring their guard down and make them more likely to trust them.
The students demonstrated how this would work in practice, by using the information they were able to find to start conversations with those they tracked down in real time.
Explaining their motivations, they said: ‘The purpose of building this tool is not for misuse, and we are not releasing it.
‘Our goal is to demonstrate the current capabilities of smart glasses, face search engines, LLMs, and public databases, raising awareness that extracting someone’s home address and other personal details from just their face on the street is possible today.’
.
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.