Ordinary WiFi can now identify you with near-perfect accuracy

ScienceDaily ^ | 8/12/2026 | Karlsruher Institut für Technologie

[Callahan]

Researchers found that ordinary WiFi routers could quietly identify people with nearly 100% accuracy, potentially turning everyday wireless networks into invisible surveillance systems. Credit: AI/ScienceDaily.com

WiFi signals could potentially be used to identify people and map their surroundings without relying on cameras or requiring the person being observed to carry a connected device.

"By observing the propagation of radio waves, we can create an image of the surroundings and of persons who are present," says Professor Thorsten Strufe from KASTEL, KIT's Institute of Information Security and Dependability. "This works similar to a normal camera, the difference being that in our case, radio waves instead of light waves are used for the recognition," explains the cybersecurity expert.

---

TOPICS: Culture/Society; Government; News/Current Events; Technical
KEYWORDS: ccp; doomedusall; ohno; paranoia; privacy; supernaughty; surveillance; technology; trusttheplan; wifi; wireless

[Callahan]

This stuff makes the Flock license plate camera controversy look like a total joke. People have no idea how powerful the AI equipped surveillance state is going to be. It’s now entirely possible to surveil not just individuals but your breathing as your chest cavity rises and falls or micro movement of your fingers. And of course future robots and other devices will have similar superhuman awareness of the environment. And for once it’s real, not a goofball conspiracy theory.

[Tench_Coxe]

Makes me wonder if we will see a ‘Butlerian Jihad’ some day.

[discostu]

The big difference between my wifi router and flock camera is my router isn’t publishing this data to a public database. Sure with enough computing power ANYTHING is a surveillance tool. But Flock was built for that.

[Red Badger]

Or a Skynet.................

[Governor Dinwiddie]

That's the reason for the massive growth in data centers. There is an exponential growth in data centers just to handle all the surveillance cameras.

[Ken Regis]

I hired six “surveillance artists“. Sure, I had to get the easels and canvases all the oil paints and brushes but every day they are painting what goes on around my property. When I get home from work, they show me what happened and I give them dinner.

[lightman]

Big difference is that I can go off the grid and d/c my router at will.

[Paladin2]

Should most WiFi routers be taken out back and Che’d?

Or should we all wear some kind of active, personal WiFi defeat device?

Can we all present as Dr. Jill?

[GingisK]

Your WIFI router does not have that ability. Only equipment specifically designed will be able to do that.

[MayflowerMadam]

“my router isn’t publishing this data to a public database.”

Are you sure? Sometimes you don’t know what you don’t know.

[GingisK]

No! An ordinary router is not equipped to do this.

[humblegunner]

There is an exponential growth in data centers just to handle all the surveillance cameras.

Thank God for you, humanity has been breathlessly awaiting
someone as bright as you and put it all together for us
with such clarity and decisiveness.

Please.. we mere mortals are so stupid... would you grant us more of your Godlike understanding?

We appeal to your superiority.. please teach us.
We are so stupid.

Help us, Obi Wan.

[KrisKrinkle]

So how come they can’t catch all the pedophiles, people who do street takeovers, rioters, illegals, porch pirates, and scammers?

[Paladin2]

“I hired six “surveillance artists“.”

You should be able to tap into Fauxci’s $tash to fund that.

Sounds kool....

[And likely more beneficial to Society and the progression ofWestern Civ.]

[GingisK]

The router is not the imaging device, it is the illumination source. The imaging devices will be new and separate.

[discostu]

I am sure. And the most important part is we DO KNOW Flock is all landing in a public database. That is the entire point of it.

[Z28.310]

“...my router isn’t publishing this data to a public database”.

Is this accurate, or a guess?
I ask because a router is, by nature, DL’ing and UL’ing data while operational. Maybe the data just stays in the provider’s unhackable database?

[Tench_Coxe]

"So how come they can’t catch all the pedophiles, people who do street takeovers, rioters, illegals, porch pirates, and scammers?"

Why were J6rs pursued and jailed, when the grossly misnamed 2020 Summer of Love rioters/organizers were not?

[ProtectOurFreedom]

Very interesting. This concept was first reported nine years ago and initially would make out objects and people. Now they have it to the point where they can identify people. But note that these people were pre-identified in the project; they cannot (yet) identify new random people.

---

The earliest well-documented reports of using WiFi like radar to image surroundings go back to October 2009, when University of Utah researchers Joey Wilson and Neal Patwari published "Radio Tomographic Imaging with Wireless Networks" and a companion paper on variance-based radio tomography, demonstrating that a mesh of ordinary WiFi-band radio signals could be used to detect and image people moving through and behind walls. That work got wide press coverage at the time, including MIT Technology Review's "Wireless Network Modded to See Through Walls" (Oct 2009) and Phys.org's "Wi-Fi signals can see through walls" (Oct 2009).

Early notable milestones:

• 2013: MIT CSAIL's Wi-Vi project (Fadel Adib and Dina Katabi) used WiFi signals to track people's motion and gestures through walls with just two antennas.
• May 2017: Technical University of Munich researchers (Holl and Reinhard) published holographic 3D imaging using ordinary 5GHz WiFi routers, letting a receiver "see" solid objects behind walls (covered in Physics Magazine).
• November 2025: KIT's KASTEL institute published newer work using WiFi beamforming feedback information (BFI) to build camera-like images of people and identify them within seconds, which is what's been recirculating in tech news (e.g., a Slashdot post from August 16, 2026, though it had already been covered back in May 2026).

The new work reported in this ScienceDaily article confirms that the KIT/KASTEL team's system, called BFId, and goes further than earlier WiFi sensing work: it doesn't just detect that a body is present, it actually recognizes which specific person it is. Key details:

• Method: Uses beamforming feedback information (BFI), an unencrypted signal that WiFi routers and devices already exchange as part of normal operation. No specialized hardware and no access to the target's own WiFi network is required — just another active WiFi device nearby.
• Accuracy: Tested on 197 participants, achieving about 99.5% accuracy, compared to roughly 82.4% for the older channel-state-information (CSI) approach on the same group. Each BFI data point contains 740 features versus 212 for CSI, giving it much higher spatial resolution.
• Robustness: Identification worked regardless of viewing angle or how the person walked.
• Important caveat — closed-set identification: The system was trained on the 197 known participants, so it matches a new sample against an enrolled roster of known people rather than recognizing a total stranger it has never seen before. It is not open-world facial-recognition-style identification of anyone off the street.
• Privacy implications: The researchers themselves flagged real privacy risks, noting that the WiFi standard lacks adequate protections against this kind of passive identification, and that mitigation attempts (like degraded sampling rates) had minimal effect on accuracy.

Sources

• Wireless Network Modded to See Through Walls - MIT Technology Review
• Wi-Fi signals can see through walls - Phys.org
• Through-Wall Tracking Using Variance-Based Radio Tomography Networks (arXiv, 2009)
• Radio Tomographic Imaging with Wireless Networks (Wilson & Patwari)
• Wi-Vi: See Through Walls with Wi-Fi Signals - MIT CSAIL
• Imaging with Your Wi-Fi Hotspot - Physics Magazine (2017)
• Researchers Use Ordinary WiFi Connections to Create Images of Nearby People and Their Surroundings - Slashdot
• Ordinary WiFi can now identify you with near-perfect accuracy - ScienceDaily
• Researchers identify people through ordinary Wi-Fi routers with 99.5% accuracy - Tom's Hardware
• Is your WiFi spying on you? - Physics World

[unixfox]

The big difference between my wifi router and flock camera is my router isn’t publishing this data to a public database.

That you know of.