• Home
  • About Us
  • Contact Us
  • Disclaimer
  • Privacy Policy
Sunday, January 11, 2026
newsaiworld
  • Home
  • Artificial Intelligence
  • ChatGPT
  • Data Science
  • Machine Learning
  • Crypto Coins
  • Contact Us
No Result
View All Result
  • Home
  • Artificial Intelligence
  • ChatGPT
  • Data Science
  • Machine Learning
  • Crypto Coins
  • Contact Us
No Result
View All Result
Morning News
No Result
View All Result
Home ChatGPT

Harvard duo modifies Meta glasses to seize strangers’ data • The Register

Admin by Admin
October 4, 2024
in ChatGPT
0
Meta Ray Ban.jpg
0
SHARES
1
VIEWS
Share on FacebookShare on Twitter


A pair of ingenious Harvard undergraduates have created what they imagine could possibly be one of the intrusive gadgets ever constructed – a wake-up name, they inform The Register, for the world to take privateness severely within the AI period.

AnhPhu Nguyen and Caine Ardayfio, who’ve collaborated beforehand on some positively explosive tasks, shared their newest mission on X within the type of a pair of Meta Ray-Bans that may try to robotically and swiftly establish anybody in view of the system’s digital camera and return an AI-generated file on them. 

Dubbed “I-XRAY” by Nguyen and Ardayfio, the mission makes use of Meta glasses to stream movies to Instagram. Faces captured from the livestream are fed via companies like PimEyes, which match the photographs to publicly obtainable ones and return the URLs. With no less than a reputation, I-XRAY can then cross-reference this knowledge utilizing people-search websites to search out addresses and different particulars – doubtlessly even partial Social Safety numbers, pieced collectively from totally different websites displaying SSN fragments.

Are we prepared for a world the place our knowledge is uncovered at a look? @CaineArdayfio and I provide a solution to guard your self right here:https://t.co/LhxModhDpk pic.twitter.com/Oo35TxBNtD

— AnhPhu Nguyen (@AnhPhuNguyen1) September 30, 2024

The server-side system doing the work, constructed by the pair in Python, spits its LLM-summarized outcomes to a cell app inbuilt JavaScript, and growth: A mini biography on anybody, obtainable immediately. Or, virtually immediately – Ardayfio advised us the app is definitely a bit gradual, and normally takes “a minute or so” to tug outcomes. 

To high all of it off, each bit of knowledge I-XRAY pulls is publicly obtainable – making this a possible open supply intelligence privateness nightmare.

All type – and a few substance, too

Utilizing a pair of sensible glasses for the mission was comparatively arbitrary, Nguyen advised us in an e-mail change, and was largely down to creating a flashy alternative that may appeal to consideration. 

“Ninety-nine p.c of the harm a foul actor may make from this instrument is unbiased of whether or not they have sensible glasses,” Nguyen defined. “Somebody may very simply, discreetly, take an image of somebody from afar – cameras have 50x zoom as we speak. They’re actually good at that.” 

Any hidden – or not-so-hidden – digital camera could possibly be used to do what the duo did, they advised us. And it would not take a lot coding know-how both: The pair solely wanted two or three days of coding, round 4 to 6 hours a day, to get the mission working, Nguyen recalled. Whereas Ardayfio has 9 years of coding expertise, and Nguyen three, that does not matter, we’re advised. 

“Anybody who can run some easy internet automations with ChatGPT can construct this,” Nguyen stated. “It is astonishing which you could construct this in just a few days – at the same time as a really naïve developer.”  

The duo would not intend to launch their code – primarily due to its potential for misuse. However they famous it was additionally initially only a facet mission that would not be match for public consumption. 

“The tech works okay,” Ardayfio advised The Register. “However it’s gradual, and never absolutely correct.” 

“Our most important objective [was] to indicate individuals what’s doable with pretty normal know-how so that folks can take their very own privateness and knowledge into their fingers,” Ardayfio added. “Unhealthy actors already know tips on how to do what we did, however we might help the great guys and most of the people be extra aware of tips on how to shield themselves.” 

Shopper Reviews’ Yael Grauer maintains an in depth listing of knowledge dealer web sites – and what must be accomplished to request data deletion – on GitHub, for many who wish to decrease their on-line presence. ®



READ ALSO

AI insiders search to poison the info that feeds them • The Register

Devs doubt AI-written code, however don’t all the time examine it • The Register

Tags: duoglassesgrabHarvardinfoMetamodifiesRegisterstrangers

Related Posts

Poison pill.jpg
ChatGPT

AI insiders search to poison the info that feeds them • The Register

January 11, 2026
Shutterstock debt.jpg
ChatGPT

Devs doubt AI-written code, however don’t all the time examine it • The Register

January 10, 2026
Shutterstock ai doctor.jpg
ChatGPT

ChatGPT Well being desires entry to delicate medical data • The Register

January 9, 2026
1767073553 openai.jpg
ChatGPT

OpenAI seeks new security chief as Altman flags rising dangers • The Register

December 30, 2025
Shutterstock 2433498633.jpg
ChatGPT

Salesforce provides ChatGPT to rein in DIY information leaks • The Register

December 25, 2025
Shutetrstock server room.jpg
ChatGPT

AI has pumped hyperscale – however how lengthy can it final? • The Register

December 23, 2025
Next Post
Generative Ai Shutterstock 2273007347 Special.jpg

Dataiku Launches LLM Guard Companies to Management Generative AI Rollouts From Proof-of-Idea to Manufacturing within the Enterprise  

Leave a Reply Cancel reply

Your email address will not be published. Required fields are marked *

POPULAR NEWS

Chainlink Link And Cardano Ada Dominate The Crypto Coin Development Chart.jpg

Chainlink’s Run to $20 Beneficial properties Steam Amid LINK Taking the Helm because the High Creating DeFi Challenge ⋆ ZyCrypto

May 17, 2025
Image 100 1024x683.png

Easy methods to Use LLMs for Highly effective Computerized Evaluations

August 13, 2025
Gemini 2.0 Fash Vs Gpt 4o.webp.webp

Gemini 2.0 Flash vs GPT 4o: Which is Higher?

January 19, 2025
Blog.png

XMN is accessible for buying and selling!

October 10, 2025
0 3.png

College endowments be a part of crypto rush, boosting meme cash like Meme Index

February 10, 2025

EDITOR'S PICK

1jhdz22u0zo8e4rkz7lqdca.png

Multimodal LLMs on Chart Interpretation

November 5, 2024
Disaster Data Center It 2 1 Shutterstock 2471030435.jpg

We Wish to Hear Your Knowledge Heart Catastrophe Tales!

February 2, 2025
Bernd dittrich dt71hajoijm unsplash scaled 1.jpg

The Hidden Safety Dangers of LLMs

May 29, 2025
1ufliw9dimri66botc9sdkg.png

Galactic Distances. How Far Are We from Alien… | by James Gearheart | Sep, 2024

September 22, 2024

About Us

Welcome to News AI World, your go-to source for the latest in artificial intelligence news and developments. Our mission is to deliver comprehensive and insightful coverage of the rapidly evolving AI landscape, keeping you informed about breakthroughs, trends, and the transformative impact of AI technologies across industries.

Categories

  • Artificial Intelligence
  • ChatGPT
  • Crypto Coins
  • Data Science
  • Machine Learning

Recent Posts

  • AI insiders search to poison the info that feeds them • The Register
  • Bitcoin Whales Hit The Promote Button, $135K Goal Now Trending
  • 10 Most Common GitHub Repositories for Studying AI
  • Home
  • About Us
  • Contact Us
  • Disclaimer
  • Privacy Policy

© 2024 Newsaiworld.com. All rights reserved.

No Result
View All Result
  • Home
  • Artificial Intelligence
  • ChatGPT
  • Data Science
  • Machine Learning
  • Crypto Coins
  • Contact Us

© 2024 Newsaiworld.com. All rights reserved.

Are you sure want to unlock this post?
Unlock left : 0
Are you sure want to cancel subscription?