South Korean electronics manufacturer LG has strongly denied allegations that its connected televisions record ambient conversations, listen while operating in standby mode, and store audio captures for subsequent transmission. The denial was issued following a 135-minute investigative report published by YouTube channel Gamers Nexus alongside Level1Techs and independent security researchers, which scrutinized the data collection and security architecture of LG smart television systems.
In its official response, LG stated that voice data is captured strictly when a user holds down the voice button on the remote control or invokes the wake word after activating Far-Field voice recognition. The company maintained that the televisions do not collect or archive background conversations under any other circumstances. Furthermore, LG explained that audio data is processed exclusively via explicit user commands and that audio is immediately deleted if the wake word is not recognized by the system.
Network Analysis and Surveillance Allegations
The independent investigation examined network traffic and software configurations, asserting that LG televisions actively track users and harvest local network data, including audio transcripts. Researchers stated they were able to record audio through an LG television while the device was in hibernation. They noted that even if a physical microphone switch on the television is disabled, auxiliary microphones connected to the system, such as those embedded in remote controls, may remain active. According to Gamers Nexus, these capabilities effectively translate into millions of monitoring devices deployed in consumer homes.
The findings suggested that recorded background noise was being archived from distances as far as 12 meters away from the television set. Researchers also alleged that unmatched wake-word interactions or ambiguous sounds could remain stored locally and transcribed as plain text. This data could theoretically be preserved on the device in temporary storage buffers in the event of a local network disconnection, awaiting future upload to corporate servers once connectivity is restored.
Security Vulnerabilities and Expert Assessment
Yash Vekaria, a computer science researcher at the University of California, Davis, emphasized the security implications of hardware longevity in home environments. Because televisions remain in residential households for years, often extending past their active software update windows, prompt security patching by manufacturers is critical. Vekaria noted that vulnerabilities could potentially allow unauthorized interception of sensitive conversations, pointing to hypothetical scenarios involving government officials discussing confidential matters near connected display units.
Discussing the potential motivations behind data retention, Vekaria raised questions regarding why voice data queued for later upload might be preserved. He suggested that such data could theoretically be utilized for logging, analytics, or speech model training. While noting an absence of direct evidence linking ambient transcripts to advertising profiles, Vekaria observed that LG operates a robust household advertising business driven by Automatic Content Recognition tracking and that its voice pipeline captures substantially more data than average consumers anticipate.
Device Discovery and Network Scanning Mechanisms
The investigation further revealed that LG televisions routinely scan home networks to discover other connected hardware, gathering internal IP addresses, nearby Wi-Fi network details, and precise location data. LG defended the practice, describing device discovery as a standard function across modern smart television platforms designed to facilitate device interoperability and media sharing.
Anna Maria Mandalari, a lecturer in communications and networking at University College London, explained that device discovery features can serve legitimate functions such as screen casting and integrating smart home products. However, Mandalari cautioned that depending on implementation methods, a television might acquire the names, addresses, manufacturers, and types of neighboring devices on a local network. This granular data enables manufacturers to map household device composition and construct detailed demographic profiles of occupants.
Corporate Advertising Practices and User Consent
Scrutiny extended to the commercial utilization of collected household data through LG Ad Solutions, the corporate advertising arm. Promotional materials highlighted by the investigative team featured representatives emphasizing granular household intelligence, including knowing the identities of occupants and the specific connected devices present within a given home environment.
In response to these findings, LG reiterated that Automatic Content Recognition, voice recognition systems, and interest-based advertising operate as distinct features. The company affirmed that each capability requires explicit user opt-in consent and can be deactivated through the television menu system. LG stated it remains fully committed to corporate transparency and explicitly rejects deceptive data collection practices.
