Judicial review claim regarding police use of automated facial-recognition technology rejected (High Court)
The High Court (Lord Justice Haddon-Cave and Mr Justice Swift) has rejected a judicial review claim in respect of how automated facial-recognition (AFR) technology had been used to date by South Wales Police (SWP), in the form of a pilot project known as AFR Locate.
The court rejected arguments put forward by Mr Edward Bridges, a civil liberties campaigner who lives in Cardiff, that SWP’s use to date of AFR Locate generally, and at two locations in the city on occasions when he had been present, had infringed his right to privacy under Article 8 of the European Convention on Human Rights. The court found that the use of AFR Locate did entail infringement of the right to privacy under Article 8(1) of the Convention of members of the public, such as those in the claimant’s position, but that the inference with the right to privacy was in accordance with the law, under Article 8(2).
The court also held that SWP’s use of AFR Locate was consistent with data protection legislation, and that the current legal regime was adequate to ensure the appropriate and non-arbitrary use of AFR Locate.
The use of AFR technology has been under the spotlight recently, with the Information Commissioner having issued guidance for police forces on the use of such technology in July 2019, and having, more recently, announced the launch of an investigation into use of live facial recognition technology at Kings Cross.
The High Court’s decision in this case does not signal the end of any questions over the use of AFR technology since, as the court acknowledged, any question of proportionality in any particular case is generally fact sensitive. However, the judgment does provide an interesting analysis of the extent to which AFR technology, used on people walking in public places interferes with their privacy, and the extent to which such interference can be deemed to in accordance with the law even if there is not specific bespoke legislative framework dealing with AFR. (Bridges, R (On Application of) v The Chief Constable of South Wales Police  EWHC 2341 (Admin) (4 September 2019)).
Caller location obligation applies to calls made from mobiles without SIM card (ECJ)
The ECJ has ruled on caller location information following a referral by a Lithuanian court in a case involving a claim for state compensation by relatives of a 17 year old girl who was kidnapped, raped and burnt alive in the boot of a car. The girl had called the single European emergency number “112” 10 times but the equipment in the answering centre had not shown the number of the mobile phone used, which prevented her being located. The claim included a submission that Lithuania had failed to ensure practical implementation of the Universal Service Directive (2002/22) (USD).
The ECJ ruled that Article 26(5) of the USD, which requires member states to ensure that caller location information is made available to authorities handling emergency calls, including calls to “112”, applied where the call was made from a mobile telephone that was not fitted with a SIM card. Article 26(5) requires that “all calls to the single European emergency call number” are covered by the obligation to make caller location information available.
The ECJ recalled its decision in Commission v Lithuania (Case C-274/07) EU:C:2008:497 (see Legal update, ECJ finds Lithuania should have ensured caller-location information available for emergency calls), in which it held that the same provision imposed on member states an obligation, subject to technical feasibility, not just to put in place an appropriate regulatory framework, but for caller location information to be actually transmitted to the emergency services.
The ECJ furthermore ruled that Article 26(5) confers on member states a measure of discretion when laying down the criteria relating to the accuracy and reliability of the caller location information, but that the criteria must ensure, within the limits of technical feasibility, that the caller’s position is located as reliably and accurately as is necessary for the emergency services usefully to come to the caller’s assistance. The ECJ relied on recitals 36 of the USD and recital 39 of Directive 2009/136 which provide that caller information is intended to improve the level of protection and the safety of users of “112” and assist the emergency services.
Finally, the ECJ also found that where an indirect causal link between an unlawful act committed by national authorities and the damage sustained by an individual was regarded by the law of the member state as sufficient to render the state liable, that must be regarded as sufficient for the purposes of rendering that member state liable for that breach of EU law.
Providing caller location information has long been technically problematic for VoIP calls (made without a SIM card. In the UK, the General Conditions of Entitlement require providers of outbound VoIP calls to recommend to mobile customers that they register and update their location information whenever accessing the service from a new location (GC A3.6(c)(ii)).
ASA reports on labelling of social media influencer advertising
The ASA has published a report on its research into consumer understanding of social media influencer advertising. Specifically, the research considered how labels (that is, advertising labels added by influencers to their social media posts) and other factors could affect a consumer’s ability to tell whether the post is in fact an advert. The research considered existing CAP guidance, recent ASA rulings, academic publications, an ASA-commissioned research and feedback from the ASA’s March 2018 public call for evidence (see Legal update, ASA calls for evidence on labelling of native and influencer advertising).
The report’s key finding is that consumers have difficulty in distinguishing all types of advertising content (even adverts posted by a brand, as opposed to an influencer) from other types of content on social media. The research also concluded that:
- The placement, visibility and wording of the labels are important factors to indicate to a consumer that an influencer’s post is an advert; for the post to be obviously identifiable as an advert the label must first be noticed, and then understood.
- The wide variety of different labels that are currently in use adds to the difficulty for consumers to identify when an influencer’s post is an advert.
The report does not indicate a change in the ASA’s approach to regulation, as set out in current guidance. In its press release announcing the publication of the report, the ASA states that its focus will be on ensuring influencers and brands are being upfront and clear with the use of #ad in their posts. However, the ASA states that this is the minimum requirement, which suggests that best practice would also involve considering how placement, visibility and wording of labels may affect consumer understanding. In terms of next steps, the report states that the ASA will further consider these findings and share them with other regulators, both domestically and internationally. On a related note, the CMA issued detailed guidance for influencers who accept payment for posts (including payment in kind) in January 2019, see Practice note, Digital marketing: an overview: Making native advertising obviously identifiable.