Technology news and discussion on TalkFuse.
Dutch data watchdog highlights safety issues with AI glasses
The Dutch Data Protection Authority (AP) has advised consumers against buying AI-powered smart glasses after tests found that they could give dangerously inaccurate advice. In one CPR simulation, a pair falsely told the wearer that emergency services had been called, while another test found glasses offering dating advice to an adult interested in a minor. The regulator said the devices are unsafe for users and bystanders in their current form.
The AP tested six models priced from under €100 to over €500, including two in simulated situations where human safety mattered. Other problems included errors interpreting medicine leaflets, toy warnings and extreme-weather alerts, as well as weak recording lights and ways to access or operate some devices without permission. The watchdog also raised concerns about captured data being processed on servers mostly outside the EU and users receiving too little information about that processing, while recognising potential uses such as helping people with visual impairments or lower literacy.
In short
- The Dutch regulator says current AI glasses pose safety risks.
- Tests found false emergency claims and mistakes with medication instructions.
- Privacy, data processing and unauthorised access also raised concerns.
Steelman: both sides, in good faith
The strongest fair case each way — we don't pick a winner.
The case for
The AP’s warning is justified because consumers may rely on glasses for urgent guidance, and false reassurance in an emergency could cause serious harm. Tests also found privacy and consent weaknesses affecting bystanders, so advising caution until makers address these risks supports public safety and informed choice.
The case against
A broad warning against buying the glasses may overlook their potential benefits, including support for people with visual impairments or lower literacy, and the tests covered only six models and selected scenarios. A thoughtful alternative is to address risks through clearer disclosures, stronger privacy safeguards and limits on high-stakes advice, while allowing people to assess devices whose uses and shortcomings are made clear.
Explained simply
AI glasses combine ordinary eyewear with cameras, microphones and artificial intelligence that can respond to what the wearer sees or hears. Some models can give spoken answers, record video or offer guidance, but their advice depends on software that may make mistakes.
The Dutch Data Protection Authority, known as the AP, oversees privacy and data protection in the Netherlands. Its concerns matter to both wearers and people nearby, who may be recorded or affected by the glasses’ advice; information gathered by the devices may also be processed on servers outside the European Union.