Last week, the YouTube channel Gamers Nexus — working with Level1Techs and independent security researchers — published something the smart TV industry has never had to face before: actual packet captures of what a smart TV does on your home network when nobody is watching. The results should change how you think about the biggest screen in your house.
The test subjects were retail LG OLEDs, including the 2025 flagship G5 — a $2,500+ television. What the researchers documented wasn't a conspiracy theory. It was on-the-wire traffic, captured with standard tools, from TVs doing exactly what they were built to do.
What the TVs Were Actually Doing
1. They cataloged every device on the network. The G5 crawled the entire local subnet and built an inventory of everything it could find — phones, smartwatches, laptops, printers. Gamers Nexus staff members who weren't part of the project had their personal devices identified by the TV. On their office network, the television enumerated 38 separate devices.
Why does a TV need to know what's on your network? LG's official explanation is device discovery — casting, content sharing, smart-home features. But that inventory didn't stay home. It fed back to LG Ad Solutions, the company's targeted advertising division.
2. They harvested nearby Wi-Fi networks to figure out where you live. This is the part most people skip past, and it's the most invasive finding. The TVs scanned for neighboring Wi-Fi network names and signal strengths — the same technique phones use to locate themselves without GPS. In firmware, researchers found configuration keys with names like enable Wi-Fi scan, location accuracy, and location poll interval.
One researcher intercepted his own TV's traffic and found it was sending location data accurate to within five miles — and he expects the BSSID-plus-signal-strength combination to provide near-pinpoint location. The TV was sending city, coordinates, and the nearest major road to an ad server 20 times per minute — while being used purely as a computer monitor, with no TV content playing at all.
3. They recorded audio in standby — and stored it offline. The researchers documented webOS TVs capturing microphone audio while the screen was off. More troubling: when they disconnected the TV from the internet, made noise near the microphone, then restored the connection — the stored audio files uploaded the moment the TV got back online. The TV was designed to cache what it heard and wait.
4. It's all an advertising pipeline. LG Ad Solutions (originally a company called Alphonso, acquired by LG in 2021) sells this data as household-level viewership datasets — a 33-million-household dataset was placed on the Databricks Marketplace in October 2025. LG executives have said on camera, "We own the glass... we own the screen and we own the data."
LG's Response — and What Changed in the Wording
LG published a statement on September 12 saying these features are "optional" and "not enabled by default," and denying that its TVs "continuously record" conversations. But Gamers Nexus caught something in the fine print: LG's earlier statement said its TVs don't collect, record, or store ambient conversations. The new statement says they don't transmit them. Store became transmit. That's not a clarification — it's a retraction dressed up as a restatement.
LG also declined to address the remote code execution vulnerabilities researchers reported in webOS — flaws that would let an attacker turn a TV into a network foothold with an always-on microphone. Those details are still under responsible disclosure.
To be fair: this isn't only an LG problem. Texas sued Hisense over similar automatic content recognition consent practices affecting 1.27 million residents, and nearly every smart TV manufacturer runs some form of ACR. LG is simply the first to get caught with packet captures this thorough.
Why This Matters More Than a Privacy Setting
Here's the part most coverage skips: your TV is the least audited computer in your house. It has a microphone, an ad business, a camera-adjacent position in your living room, and a permanent network connection. It receives updates on the manufacturer's schedule. Almost nobody ever inspects what it sends over the wire.
And on most home networks, the TV sits on the same flat network as your work laptop, your phones, your security cameras, and your files. If a television gets compromised — remember, researchers found remote code execution flaws — the attacker doesn't just get the TV. They get a launching pad toward everything else you own.
The toggle in the TV's settings menu helps, but it's buried six menus deep, and the researchers documented dark patterns in the consent flow — including a "do not sell" toggle whose off position sits next to the word "hide" in a way that reads as on. Settings alone aren't a security strategy.
How to Actually Lock This Down
The nuclear option: Never connect the TV to the internet. Feed it with a Roku, Apple TV, or Fire Stick and leave the TV's Wi-Fi disabled. You lose nothing except the TV's own ads-and-tracking apps, and you gain a TV that can't phone home. This is what privacy-conscious users have done for years.
The practical option — and what we recommend: Put the TV (and every other smart device) on a separated network segment. In networking terms, that's a VLAN — a virtual wall inside your router. Done properly:
- Your TV can reach the internet for Netflix updates, but cannot see or scan any other device you own
- Your phones, laptops, and work equipment don't share a network with the least-trusted device in the house
- A compromised smart device can be wiped out by an attacker without ever touching anything that matters
- If you want the TV to never send data anywhere, block its DNS entirely — it works fine as a dumb screen
Business-grade routers and mesh systems (Ubiquiti UniFi, Omada, pfSense/OPNsense, even many ISP gateways' guest-network modes) support this. The difference between a flat network and a segmented one is the difference between one compromised device meaning nothing and one compromised device meaning everything.
The Bottom Line
A television used strictly as a computer monitor was caught beaconing street-level location data to an advertising company twenty times a minute. That sentence should end the debate about whether smart TVs are privacy-safe by default.
You don't have to throw the TV out. But it should live on its own island — and if you're not sure whether yours does, that's worth finding out. We run network privacy audits for Knoxville-area homes: we'll tell you exactly what's talking to what on your network, lock down the devices that shouldn't be trusted, and set up proper isolation so one bad device can never take down the rest.
Concerned about what's on your network? Get a free network assessment quote or learn more about our network security services.
Sources & Further Reading
- GamersNexus — "216,000,000 Spy TVs" investigation (September 2026), including the packet captures showing per-minute location beacons and LAN device cataloging: youtube.com/@GamersNexus
- Level1Techs — technical analysis of the LG telemetry data streams and network behavior: level1techs.com
- The Verge — coverage of LG's privacy policy wording change from "store" to "transmit": theverge.com
- hwbusters.com — independent reporting on the smart TV data collection findings: hwbusters.com
- Malwarebytes Labs — analysis of smart TV privacy risks and what the LG findings mean for home networks: malwarebytes.com/blog
- PPC Land — reporting on LG's advertising division and household-level viewership data sales: ppc.land
- LG's official statement — company response to the investigation, as quoted in the coverage above
We recommend reading the original investigations — the packet-capture evidence is worth seeing firsthand.
