Data & privacy

Smart glasses and invisible recording: where privacy law stops and the real problem begins

The ICO has warned that wearable cameras make it harder to know when personal data is being captured. Data protection law does not prohibit every personal recording, but wider sharing, organisational use, biometric analysis and targeted harassment can change the position substantially.

In this article

The device is visible, but the collection may not be

People have learned to recognise a phone held up to take a photograph and signs warning that CCTV is operating. Wearable technology weakens those cues. A camera can point in the same direction as the wearer, record continuously or send material for remote processing without the people nearby understanding what is happening.

The ICO also makes an accessibility point. Visual indicators and signs have never given equal transparency to people with no or low vision. As devices become less obvious, the same failure becomes a broader public problem: how can someone avoid, question or object to processing that they cannot identify?

The questions are wider than recording alone

  • Is the device capturing video, audio, location or a combination of them?
  • Is the material kept locally, uploaded to a platform or processed by an AI service?
  • Can people tell when collection begins and ends?
  • Is facial or biometric analysis being used to identify individuals?
  • Is the recording genuinely personal, or is it being shared publicly or used by an organisation?
  • Is a particular person being followed, targeted, humiliated or placed at risk?

Personal use can fall outside data protection law

UK data protection law contains an exemption for processing carried out by an individual in the course of a purely personal or household activity. The ICO says this means the law does not prevent people taking photographs or videos for personal use. A bystander cannot assume that UK GDPR applies simply because their image was captured.

The exemption depends on use, not the brand of device

A pair of smart glasses does not automatically create a data-protection breach, and a smartphone does not automatically avoid one. The purpose, context, audience and subsequent use of the personal data are more important than the physical form of the camera.

Public sharing can change the analysis

The ICO says a recording posted so widely online that anyone can see it is unlikely to remain purely personal use. Data protection considerations may then apply. The boundary is contextual: sharing privately with a small circle is different from building an audience, monetising clips, systematically recording strangers or publishing material for a commercial purpose.

How the context changes the issue

SituationLikely questionPossible route
Tourist records a personal clipIs this purely personal or household activity?A polite request or venue policy may be more relevant than an ICO complaint.
Creator publishes strangers to a public audienceHas the use moved beyond the household exemption?Ask for removal, use the platform process and consider a data-protection complaint.
Business staff use wearable camerasWhat lawful basis, notice, retention and access controls apply?Complain to the organisation and then the ICO if unresolved.
Software identifies facesIs biometric data being processed to uniquely identify people?Stricter UK GDPR conditions and a documented impact assessment may be required.
Wearer repeatedly targets one personDoes the conduct amount to harassment, stalking, intimidation or another wrong?Preserve evidence and use the route matching the conduct, which may include the police.
Recording occurs in a restricted venueWhat contractual, safety or confidentiality policy applies?Raise it promptly with the venue, employer, school or service provider.

Recording in public is not automatically unlawful

There is no general rule that every photograph or recording made in a public place requires the consent of every person visible. That does not make every use acceptable. Audio capture, expectation of privacy, confidential settings, children, targeted conduct, publication and automated analysis can all create additional legal and safeguarding questions.

Biometric processing raises the stakes

A photograph of a face is personal data when a person is identifiable, but it is not automatically special-category biometric data. That higher classification generally becomes relevant where specific technical processing is used for the purpose of uniquely identifying a person. Facial recognition or matching can therefore create obligations beyond ordinary recording.

What to do if you believe you were recorded

Match the evidence to the harm

  1. Record the date, time, location, device, wearer and what indicated that recording was taking place.

  2. If it is safe, ask whether recording is active, its purpose, where the material goes and whether you can be excluded or deleted.

  3. Identify any organisation involved, such as an employer, shop, venue, school, transport provider, platform or device service.

  4. If material appears online, preserve the URL, account name, screenshots, audience size and publication date before reporting it.

  5. Use the platform's privacy, harassment or non-consensual imagery reporting route and keep the decision.

  6. Complain to the responsible organisation where its processing is involved, asking about lawful basis, retention, recipients and deletion.

  7. Contact the police promptly where the behaviour involves threats, stalking, sexual misconduct, blackmail or an immediate safety risk.

Deletion is not automatic in every case

An organisation may have a legitimate reason to preserve footage, including evidence of an incident or a legal claim. Ask for restriction and preservation where the material is disputed, rather than demanding immediate deletion and then discovering that evidence needed for your complaint has disappeared.

Do not stigmatise every wearer

The ICO's Chief Executive, who is severely visually impaired, highlighted how smart glasses can support navigation, access to information and independence. The regulator has also worked with technology intended to assist people living with dementia. Responsible assistive use should not be treated as suspicious merely because other users bypass safeguards or behave intrusively.

Good product design should make responsible use easier. That can include reliable capture indicators, accessible explanations, clear controls, limited default collection, secure processing and barriers against disabling safety features. Privacy and accessibility should be designed together.

Why this matters to ConsumerWise readers

The story exposes a gap between public expectation and regulatory reach. A person can experience an intrusive recording without the individual wearer necessarily falling within UK GDPR. The practical response is to identify the actor, use and harm, then choose among data protection, platform rules, venue policy, safeguarding, harassment law or police action rather than assuming the ICO can solve every case.

The bottom line

Smart glasses make recording easier and less visible, but the correct response is neither a blanket ban nor the claim that existing law covers everything. Identify what was captured, who processed it, how it was used and whether the conduct created a separate harm. That is the route to a complaint capable of reaching the right decision-maker.

Official sources

Jack Whittle

About the contributor

Jack Whittle

Founder & Editor, ConsumerWise

Jack Whittle is the founder and editor of ConsumerWise. His background spans HR and policy, complaints and dispute resolution, digital publishing and consumer information. He researches and translates consumer law, regulation and official guidance into practical guides, tools and resources.

ConsumerWisehttps://consumerwise.org.uk/news/smart-glasses-invisible-recording-privacy-law/