Apple says its glasses will have bigger, brighter lights so people know when you're recording them. Here's the thing: nobody notices the light. The BBC tested Meta's version — not a single woman interviewed spotted it. Two Harvard students proved you can pair smart glasses with facial recognition to instantly get anyone's name, home address, and phone number just by looking at them. No light stops that. Apple's "more prominent LED" is a fig leaf on a product category that fundamentally cannot protect the people being filmed. Apple says it protects your biometric data on wearables. Its Vision Pro headset tracks your eyes so precisely that researchers found it can reveal your health conditions, emotions, and sexual preferences — and you cannot turn that tracking off because the device literally won't work without it. A vulnerability called GAZEploit let hackers figure out what you were typing by watching your eye movements. Apple blocks other companies from accessing this data, but gives itself full access — what NYU flagged as antitrust hiding behind privacy. The company that restricts others' access to your data while keeping full access for itself is not a privacy champion. It's a monopolist with better marketing.
What they claim: Apple plans "more prominent" camera indicator lights on the N50 glasses to signal recording to bystanders, positioning this as superior to Meta's approach
What we found: The fundamental problem is not the indicator light — it is that bystanders cannot meaningfully consent to being recorded. The BBC found multiple videos showing how Meta's LED can be covered or disabled; none of the women who spoke to them noticed the indicator. Harvard students built I-XRAY using Ray-Ban Meta glasses to instantly identify strangers via facial recognition — name, address, phone number — just by looking at them. A National Law Review analysis found "existing laws are really not sufficient to protect us from the risks" of wearable cameras.
What they claim: Apple positions its wearable ecosystem as privacy-protective, restricting third-party access to eye-tracking and camera data with "privacy by design"
What we found: Research shows Apple Vision Pro's eye-tracking data can reveal health conditions, mental states, sexual preferences, and emotional responses — yet this biometric collection cannot be opted out of. The GAZEploit vulnerability allowed hackers to decrypt data entered on Vision Pro's virtual keyboard by analysing eye movements. Apple restricts developer access to cameras but exempts its own apps — what NYU flagged as a potential antitrust issue disguised as privacy protection.
What they claim: Apple markets itself as the privacy-first alternative with campaigns like "What happens on your iPhone, stays on your iPhone" and positions its smart glasses as the responsible alternative to Meta
What we found: Apple paid $95 million to settle the Lopez v. Apple class action after Siri was caught recording private conversations through unintended activations and sending those recordings to subcontractors in Ireland — workers listened to up to 1,000 Siri recordings per day. A separate Illinois BIPA case (Zaluda v. Apple) certified a class of 2.6-3.9 million Siri users in January 2026, with potential exposure of $1,000-$5,000 per violation. Apple's smart glasses will use Siri as the core interface.
What they claim: Apple emphasises on-device processing as a core privacy advantage, stating the N50 glasses "set themselves apart from Meta and Google's cloud-dependent models"
What we found: Apple's own lawyers stated in court: "Given Apple's extensive privacy disclosures, no reasonable user would expect that their actions in Apple's apps would be private from Apple." Proton's analysis found Apple's own apps send analytics even when users opt out. Apple collects device data, location, App Store searches, and viewing habits for targeted ads. The glasses' Visual Intelligence feature will generate vast new streams of contextual data about everywhere users look.