Meta remains silent on whether Ray-Ban smart glasses photos are used to train AI

Meta is refusing to clarify whether images captured by its AI-powered Ray-Ban smart glasses will be used to train its AI models, raising concerns about privacy.

The glasses, which feature a discreet camera, can take photos both intentionally and passively based on voice commands or triggered by keywords like "look." Despite the large volume of images collected through these features, Meta has not committed to keeping these photos private.

When TechCrunch asked Meta if it plans to use these photos for AI training, as it does with public images from Instagram and Facebook, the company avoided a direct answer. Anuj Kumar, a senior director at Meta, stated that the company is "not publicly discussing that," and spokesperson Mimi Huggins echoed that Meta isn't sharing information on the matter either way.

This lack of transparency is particularly concerning given the glasses' new AI capabilities, which will soon include a real-time video feature that streams images to an AI model to answer questions about the user’s surroundings. These features could result in the capture of numerous images that users might not be aware are being taken, such as when scanning a room or closet.

Meta’s ambiguous stance contrasts with other tech companies like Anthropic and OpenAI, which explicitly state that they do not train their AI models on user inputs. And Meta remains silent on whether personal images captured through its smart glasses are also used in this way.