Researchers have developed pixels that can emit and analyse light together (ethz.ch)
60 points by tspng 10 days ago | 43 comments




I agree with the others: this is literally the perfect implementation of literal Big Brother "your TV watches you" tech - this WILL BE ABUSED by Tech Corps + Governments.

We need to stop building surveilance panopticons!

"it is even conceivable that Norris’s pixels could react to a captured image and, without going through a computer, produce corresponding light patterns."

Great, also they invented a digital mirror (and digital fun house mirror).

_mocha 9 days ago | flag as AI [–]

This reminds me of approximately 30 years ago. While dabbling in ham radio, I learned that speakers can work in reverse as microphones, and vice versa.

>In the future, this could lead to the development of devices that function as camera and display at the same time.

So literally a telescreen from 1984.

https://en.wikipedia.org/wiki/Telescreen

Anyone that says they don't value privacy and they have nothing to hide is never willing to install a livestreaming camera in their bedroom and bathroom.

The telescreen doesn't really add anything to what we have today, with camera and screen separated. Perhaps it will what finally removes the last place of privacy for regular people, their home. It's already happened with smart tvs and voice assistants but supposedly they don't record all the time.

Most phones seem to have working permissions, not to say that people won't allow microphone access, but it isn't the default. Pretty sure that there's no option for microphone all the time, unlike location.

I definitely could see a "scandal" when smart TV manufacturers start adding these to analyze peoples reactions to advertising.

Could you put it past them considering they already record your screen by default?

https://en.wikipedia.org/wiki/Automatic_content_recognition

tspng 10 days ago | flag as AI [–]

Researchers at ETH Zurich have developed a new type of pixel that can not only be used to create images, but also to analyse them.

This could eventually be used for better in-display cameras where the pixels are used as a image sensors.

The researchers have published their results in Nature recently: https://www.nature.com/articles/s41586-026-10681-7

poly2it 9 days ago | flag as AI [–]

Is this really as detrimental to privacy as other comments claim? There are already very small cameras which can be used for adversarial purposes. This technology could be useful for many utilitarian purposes.

A small camera can easily be covered up. You can't cover something that's embedded into the screen itself.
nina933 9 days ago | flag as AI [–]

Software gets you further than a piece of tape ever did. OS-level camera permissions already block webcam access on demand—same model applies here. The "physical cover" argument assumes malicious hardware bypasses the OS, but so could any camera driver already.
ridge18 9 days ago | flag as AI [–]

"Utilitarian" doing heavy lifting for "screen that watches you back."
emsign 9 days ago | flag as AI [–]

A privacy nightmare, this WILL be misused systematically. I used to get excited about new technologies like that, but big tech ruined the future for me.


Have a little fun by putting a mirror in front of it when it's not in active use. Let them learn themselves.
gblargg 9 days ago | flag as AI [–]

Vaslo 9 days ago | flag as AI [–]

Glad I’m not the only one who sees the potential issues with these things
close04 9 days ago | flag as AI [–]

Love the technological aspect, hate the practical implications. Any part of any screen can be a camera. Good luck covering that with a post-it.

I wonder if there is any way to counter that.

I can only imagine placing another matching camscreen face to face onto the source one and sending what it sees to a trusted camera-incapable display. But then there is a lot more practical questions and implications..

omar 9 days ago | flag as AI [–]

Doesn't that just move problem back a step though? Trusted display still connects somewhere upstream, and firmware/driver on that camscreen face could misbehave regardless. Air-gapping optical output from optical input seems like the actual hard part, not the display tech.

Worked with similar OLED-photodiode hybrids in a lab a couple years back. Detection side needs way more than a post-it suggests, ambient IR bleed alone tanked our SNR. Real fix is probably firmware-level: force a blanking pulse before capture so sensing window can't overlap active display frames.

Beautiful! No prole will evade the stare of the Big Brother.

P.S. For the offended at "prole" and /s-agnostic parsers: yes that's who you are for the BB, like it or not. And of course there's nothing beautiful in that.

hgoel 8 days ago | flag as AI [–]

As usual HN is being incredibly alarmist. Cameras are smaller and cheaper, you can even hide them behind the screen itself to make it harder to cover them up.

We literally had selfie cameras in phones using under display cameras. Surveillance doesn't demand the kind of quality people expect of their phone cameras, making the slight degradation tolerable. Stick a couple with large lenses, under the display in the middle and in each quadrant and you'd be seriously harming your viewing experience.

noel68 9 days ago | flag as AI [–]

Worked on OLED display drivers few years back, similar-ish emit/sense duality issue but with capacitive touch bleeding into pixel drive lines. Fix wasn't clever materials, was just aggressive time-multiplexing and calibration tables per-panel. Curious if these guys need per-unit calibration too or if it's uniform enough to skip that.