Talk:LG Television sale of personal data

Does the opt-out genuinely work?

I've just bought a B5 OLED from LG as the cheapest possible OLED option new, locally. Found my way to the opt-out before anything else, ensured live plus was off, etc. but even then still did not connect it to my home network. That is, until I found a solid advertising/tracking URL blocklist to add to my router to ensure that any attempts to continue phoning home are blocked. However, after setting that list up and finally connecting it (by ethernet or wifi) the TV continues to insist it is offline and prevents access of ANY online features. This is immediately suspect to me, as the only endpoints blocked were for advertising and tracking. If it truly was not attempting to do so after the opt out, access to those URLs should not even be attempted, let alone required before the TV will acknowledge an internet connection. I have the necessary hardware to run it through a switch before my router to run wireshark in-between and see exactly where it's trying to communicate, but that will require cutting another 50ft line of cat5e, so before I do that I was curious if anyone else has, and what the results were.

Additionally, I've heard WebOS has a developer mode that can enable sideloading and possibly jailbreaking. Has anyone gone through the effort of modding the OS such that it no longer includes personal data sale features, and from there compared the communications it sends out to both an unmodified and opted-out version? 65.182.79.52 05:42, 5 January 2026 (UTC)

This is partly answered now by two 2026 investigations, though they do not fully agree with each other.

Gamers Nexus captured traffic from an LG G3 and reported that a set used only as an HDMI display, with no agreements accepted at all, still contacted eight LG advertising and customer data endpoints over an hour and sent 38 DNS queries to LG domains. That is broadly consistent with what you describe, in that the set appears to want those endpoints regardless of consent. In fairness they also reported that no ACR data appeared in that no-agreements state.

RTINGS.com reached a different conclusion on the narrower question of ACR specifically. Testing 15 devices with Wireshark before and after withdrawing consent, they reported that on an LG G5 the identifiable ACR traffic "stopped or changed dramatically" once consent was withdrawn. Both findings are now in the article, presented as differing rather than one being treated as settled.

Two practical notes on your Wireshark plan. Most payloads are TLS encrypted, so a capture will give you destination servers, timing and volume but not the contents without a MITM certificate; RTINGS ran into the same limit. And on the TCL set they tested they could not separate ACR traffic from ordinary telemetry at all, which they noted is precisely what makes DNS blocklists unreliable for this purpose.

Worth flagging that under Consumer Rights Wiki:Inclusion guidelines and the no original research rule, your own capture results would not be citable in the article itself. They would still be genuinely useful here on the talk page, and if a published source later documents the same behaviour it becomes usable.

On WebOS developer mode, Gamers Nexus decompiled LG firmware with help from the TV jailbreaking community, so that community is probably the most productive starting point for the comparison you describe. Nnayak (talk) 08:02, 13 September 2026 (UTC)

2026 update following the Gamers Nexus investigation

Louis Rossmann asked viewers to help overhaul this article in a video on 13 September 2026, so others may be working on it too. Here is what I have added today, to save anyone repeating it.

Added:

• A 2026 investigation section covering the Gamers Nexus series of July to September 2026. The packet capture figures are quoted from the video's own captions, and I have left the timestamps in HTML comments next to each one so they can be checked without watching the full two hours.
• An LG's response section built on LG's statement of 12 September 2026. Worth noting for this article in particular: LG says there that ACR-related information "may also be shared in ways that may be considered a 'sale' or 'sharing' of personal information under applicable privacy laws".
• A Regulatory action section. The Texas Attorney General's case against LG was settled on 11 May 2026 by an Agreed Final Judgment in Bell County, Cause No. 25DCV358507. That was not recorded anywhere on the wiki before today.

Corrected:

• The opt-out section previously said the process took a "27-click process" and then listed four steps. The Markup's 27-click figure is its count for turning off Live Plus, which is a different control from the "Do Not Sell My Personal Information" toggle. These are now separated. The Texas petition's separate "40+ clicks" figure is in the Regulatory action section.
• Removed wording that attributed intent or stated a legal conclusion in the wiki's voice, per Consumer Rights Wiki:Editorial guidelines.
• The Cargo description was 159 characters against the documented 150 character limit.

Two things I deliberately did not add, and why:

• Gamers Nexus demonstrated microphone and camera capture, but stated it required exploiting software vulnerabilities. I have kept that caveat in the text. The article should not say LG records conversations, because the source does not say that.
• Rossmann also mentions LG washing machine terms. The policy text he reads on screen says LG "share" personal information, not "sell". The "sells" framing is his own stated inference, so that belongs on LG ThinQ data collection with attribution rather than here.

Still open, if anyone wants to pick it up:

• RTINGS.com tested 15 devices in 2026 and reported that ACR traffic stopped or changed markedly on an LG G5 once consent was withdrawn. That partly differs from the Gamers Nexus finding and both should be presented, per Consumer Rights Wiki:Verifiability on conflicting sources. I have not added it yet because I have not read the article closely enough to quote it.
• LG also gave a statement to Tom's Hardware on 9 September 2026 saying the claims "are not true". That predates the newsroom statement and should open the response section. I have not added it because I have not verified the quotation against the source.
• The article is still scoped to the LG G3 OLED in Cargo, but the sources now cover several models. Widening it seems right, though I would rather agree that here first than change it unilaterally.

Nnayak (talk) 07:55, 13 September 2026 (UTC)