Smart TVs are actively scraping and transmitting user data to train AI models, becoming nodes in a sprawling data collection network that most owners don't realize exists.
Smart televisions installed in millions of homes are being repurposed as data collection devices, feeding information into AI training systems without explicit user consent or awareness.
The devices capture viewing habits, household patterns, and behavioral data, which is then aggregated and sold to AI companies developing large language models and other machine learning systems. This occurs alongside traditional ad tracking, creating a dual revenue stream for TV manufacturers.
Security researchers note that the infrastructure enabling this practice is largely invisible to consumers. Users typically accept terms of service without understanding the data pipeline their devices participate in.
The practice raises questions about consent, data ownership, and regulatory oversight. Manufacturers argue data collection funds free or affordable device models, while privacy advocates warn this represents a fundamental shift in how consumer devices operate.
The issue gained traction in tech communities after detailed analysis of device behavior, with discussions emerging on platforms like Hacker News regarding implications for privacy and regulatory frameworks.
A zero-day vulnerability called StyleSmuggler is being actively exploited across all versions of Magento and Adobe Commerce to install Linux backdoors on compromised systems.
A phishing-as-a-service platform called BigBear 2.0 has successfully circumvented multi-factor authentication defenses to compromise more than 5,000 Microsoft 365 credentials across 258 organizations.
Cryptocurrency hardware wallet maker Trezor revealed that an August data breach at logistics provider ShipMonk impacts 81,000 customers total, with an additional 67,000 U.S. customers newly affected.
Security researchers discovered that LG smart TVs continue recording audio and scanning local networks even when the display is powered down. The findings raise concerns about user privacy and device security.