Reports circulating online allege that some televisions continue collecting and transmitting data in standby or with limited connectivity. The underlying question — how much a smart TV observes and when — is real, even where specific claims remain unverified.
Key takeaways
- Claims that certain LG televisions gather data while switched off or disconnected have circulated widely on technical forums, but the technical details behind them have not been independently confirmed in any account that can be verified here.
- Modern smart televisions run full operating systems that remain partially active in standby, which makes “off” a far less precise state than it was on older sets.
- Automatic content recognition, a technique that samples what appears on screen to identify programmes and advertisements, is a well-documented feature of the smart TV industry generally.
- A device with no network connection can still buffer data locally and transmit it later, which is why “offline” does not automatically mean “not collecting”.
- Regulators in several jurisdictions have examined connected television data practices before, and the legal questions turn on consent, disclosure and the definition of personal data.
What is actually being claimed
The trending discussion concerns allegations that televisions from a major manufacturer continue to gather information about users even when the set appears to be switched off, or when it is not connected to the internet. The specific technical evidence behind these claims — what was captured, using which tools, on which models and firmware versions — is not something that can be verified from the available signal, and this article does not assert that the claims are accurate.
What can be discussed with confidence is the architecture that makes such claims plausible enough to spread. A contemporary smart television is a networked computer with a screen attached. It runs an operating system, maintains background services, applies firmware updates and, in many product lines, participates in advertising and measurement systems. Those capabilities do not switch off cleanly when a user presses the power button on a remote.
Why this is surfacing now
Discussions of this kind tend to resurface whenever someone posts packet captures, firmware analysis or router logs showing unexpected traffic from a consumer device. The technical community treats such posts as testable claims, and the comment threads typically split between people reproducing the finding, people identifying benign explanations, and people pointing out that the behaviour was disclosed in a privacy policy nobody reads.
There is also a broader context. Connected televisions have become a significant advertising channel, and the economics of the category reward manufacturers who can sell televisions cheaply and recover revenue from data and advertising afterwards. That business model has been described openly by industry participants for years. When a specific allegation appears, it lands in an audience already primed to find it credible.
The background a newcomer needs
Three technical concepts explain most of what is at stake.
The first is standby. Televisions do not fully power down when turned off with a remote. They enter a low-power state that keeps enough of the system alive to respond to the remote, to wake for scheduled recordings or updates, and in some cases to listen for wake-on-network commands or voice triggers. The amount of activity permitted in standby is a design choice, and it varies by model and configuration.
The second is automatic content recognition, usually abbreviated ACR. This technique periodically samples frames or audio from whatever the television is displaying, converts them into compact fingerprints, and matches them against a reference database. It allows a manufacturer to know what a household watches, including material arriving through an external device such as a games console or set-top box, which the television would otherwise have no visibility into. ACR is widely deployed across the smart TV industry and is typically presented to users during setup under a marketing name rather than a technical one.
The third is buffering. A device without network access can still write observations to local storage and upload them when connectivity returns. This is ordinary engineering practice for telemetry systems of all kinds, and it is the reason a claim of the form “it collected data while offline” is not self-contradictory. Whether it happened in any particular case is a separate question from whether it is technically possible.
Who is affected, and how
The people most directly affected are households that own connected televisions, which in many markets now means most households buying a new set. It is increasingly difficult to buy a large television that is not a smart television, so the choice to avoid the category is not really available at the point of purchase.
The exposure is not evenly distributed. Someone who completes the initial setup wizard by accepting all defaults, connects the television to home Wi-Fi and signs into a manufacturer account has a considerably larger data footprint than someone who declines optional terms, never connects the set to a network and uses it purely as a display for an external device. Viewing history is also more sensitive than it first appears: what a household watches can indicate political leanings, health concerns, religious observance, household composition and daily routine.
There is a secondary group: anyone on the same network. Devices that behave unpredictably on a home network are a concern for people running small businesses, home offices or networks that carry sensitive traffic, independent of any privacy question about viewing habits.
Where informed people disagree
The disagreements are genuine and worth stating fairly.
One dispute concerns interpretation of network traffic. A television contacting a manufacturer’s servers in standby may be checking for firmware updates, synchronising time, or maintaining a connection for a legitimate remote-wake feature. Observers looking at encrypted traffic from outside the device can often see that communication occurred and roughly how much, but not what it contained. Reasonable engineers disagree about how much can be inferred.
A second dispute concerns consent. Manufacturers generally argue that data collection is disclosed and that users agree to it during setup. Critics argue that consent obtained through a lengthy screen on a television remote, at the moment someone wants to start watching, is not meaningful, and that defaults matter more than disclosures. This is as much a legal and design argument as a technical one.
A third concerns proportionality. Some hold that content recognition applied to a household’s viewing is categorically different from ordinary product telemetry, because it observes behaviour rather than device health. Others treat it as a routine part of an advertising-funded product that keeps hardware prices down. Regulators in different jurisdictions have taken different positions.
The practical implications
For an individual, several mitigations exist, with different trade-offs.
The most complete is to keep the television off the network entirely and drive it from a separate streaming device or computer, which moves the trust decision to a device that is easier to replace. This costs smart features and, on some models, firmware updates that include security fixes.
A lighter approach is to work through the television’s privacy settings and decline the options relating to viewing information, personalised advertising and content recommendations, which on most platforms are separate toggles rather than one. Settings can be reset by firmware updates, so they are worth re-checking periodically.
Network-level controls — blocking known telemetry domains at the router or DNS resolver — offer another layer, though they can break legitimate functionality and require maintenance as endpoints change.
For organisations, the implication is that televisions in meeting rooms and public areas are network endpoints and should be treated as such: segmented onto a guest or IoT network, inventoried, and included in patching schedules.
What to watch next
Watch for independent technical verification. Claims of this kind become substantially more credible when a second party reproduces the finding on named models and firmware versions with a published method. Absence of reproduction after a claim circulates widely is itself informative.
Watch for a manufacturer response describing what the observed behaviour actually is. Vendors often clarify that particular traffic serves a specific function, and such statements are checkable.
Watch for regulatory interest. Connected television data practices have attracted attention from consumer protection and data protection authorities in multiple jurisdictions, and formal inquiries tend to produce more reliable documentation than forum threads.
Finally, watch the direction of default settings on new models. Whether privacy-relevant features ship enabled or disabled by default is the single clearest indicator of where the industry is heading, and it is observable by anyone setting up a new television.
Frequently asked questions
Does turning off a smart TV stop it collecting data?
Not necessarily. Pressing the power button on a remote usually puts a television into standby rather than powering it down completely. In standby, parts of the system stay active so the set can respond to the remote, check for updates or support wake-on-network features. Whether any given model transmits data in that state depends on its design and settings. Unplugging it from mains power is the only certain way to stop all activity.
What is automatic content recognition on a television?
Automatic content recognition samples what is displayed on screen or played through the speakers, converts it into a compact fingerprint, and compares it against a reference database to identify the programme, film or advertisement. It lets a manufacturer see what a household watches, including content arriving from an external device such as a console or set-top box. It is common across the smart TV industry and is usually presented under a product-specific name during setup.
Can a television collect data if it is not connected to the internet?
A device with no network connection cannot transmit anything at that moment, but it can record observations to local storage and send them later once connectivity is restored. This is standard practice for telemetry systems generally. So a claim that a device collected information while offline is not technically contradictory, though whether it occurred in any specific case requires evidence about that device and firmware.
How do I check what my television is sending?
The most accessible method is to inspect traffic at the network level: many routers log connections by device, and DNS-filtering services show which domains a device queries. This reveals where a television is connecting and how often, but not the contents of encrypted traffic. More detailed inspection requires network capture tools and technical knowledge, and interpreting the results reliably is harder than collecting them.
Is smart TV data collection legal?
It depends on the jurisdiction and on how consent was obtained. Data protection regimes generally require that collection be disclosed and, for some categories, that users agree meaningfully. Manufacturers typically rely on terms accepted during setup. Critics dispute whether such consent is meaningful given the interface. Regulators in several jurisdictions have examined connected television practices, and outcomes have varied.
What is the simplest way to reduce a smart TV’s data collection?
Not connecting the television to a network at all is the most complete option, using a separate streaming device instead. Short of that, go through the privacy menu and decline options covering viewing information, personalised advertising and recommendations, which are usually separate toggles. Re-check these after firmware updates, since settings occasionally change. Declining optional terms during initial setup also reduces what is enabled by default.
Sources and further reading
- Consumer protection and data protection authorities in multiple jurisdictions, which have published findings and guidance on connected television data practices.
- Academic and independent research groups that have measured network behaviour of consumer smart televisions using traffic analysis.
- Manufacturer privacy policies and settings documentation, which describe the categories of data collected and the controls offered.
- Technical community forums and mailing lists, where reproducible network captures and firmware analyses are posted and debated.
Surfaced from the hackernews signal “smart TV privacy claims”. AI-assisted draft, editorially reviewed.
