GoKawiilResearcher finds Samsung Tizen TVs running AlexNet model to identify on-screen content
A researcher identified as log4j reported that Samsung Tizen devices run a TensorFlow Lite model, file-named 'alex.tflite' and based on AlexNet, repeatedly processing 960x540 screen captures with INT8 quantization to try to detect what game or content a user is viewing. The researcher says the model performs nearest-neighbor comparisons against a database, often fails to produce accurate results, and appears to store its output before any user consent prompt is shown.
GoKawiil's interpretation of the reporting above, not reported fact.
If confirmed, continuous on-device image analysis before a consent gate would raise privacy questions about how Samsung Tizen devices handle screen content and user data, since processing reportedly occurs regardless of whether a user has agreed to data collection. The researcher's findings, shared via live-tweeted technical analysis, suggest possible inefficiency and inaccuracy in the feature as well, though Samsung has not publicly responded to these specific claims.
- A TensorFlow Lite model based on AlexNet reportedly runs on Samsung Tizen TVs to analyze screen captures.
- The researcher claims the analysis occurs before any consent screen is presented to users.
- The findings come from independent research shared on social media, not from an official Samsung disclosure.
Source: twitter.com, 2026-10-25
Published there as: “Samsung running a TensorFlow lite model on 960x540 image to track you”
Read the original report → The summary and analysis above are GoKawiil's own, written from reporting by the source above. Facts and quotes belong to the original publisher.