Onward data transfers
On 13 November, the European Data Protection Board approved the use of other models of contractual clauses for onward transfers of personal data based on updates made by the European Commission. The Model Contractual Clauses (MCCs) developed by the Council of Europe (CoE MCC) and the ones developed by the Ibero-American Data Protection Network (RIPD MCC) can now be used for onward data transfers received under the EU Standard Contractual Clauses (SCCs).
This means that, under the conditions set by the Commission, an appropriate level of protection of individuals as guaranteed by the European data protection law may be provided when data is onward transferred outside the EU/EEA based on these clauses.
Stay up to date! Sign up to receive our fortnightly digest via email.
Particular attention should be paid to compliance with the purpose limitation principle. In addition, where sensitive data are subject to onward transfers, entities should ensure that such data continue to benefit from the higher level of protection applicable to those categories of data.
The EDPB sees these additional instruments as fostering future interoperability of various model contractual clauses. Onward data transfers based on either the CoE MCCs or the RIPD MCCs may take place if a corresponding module is available. For instance, as of the date of the statement, the RIPD MCCs can be used only for onward transfers from controller to controller and from controller to processor, but not from processor to processor.
In parallel, the EDPB has adopted the final version of the recommendations on the application for approval and on the elements and principles to be found in Processor Binding Corporate Rules (BCR-P). This transfer tool can be used by a group of undertakings or enterprises to transfer personal data outside Europe to processors within the same group.
Digital Fairness Act
On 11 November, the European Commission presented a proposal for a Digital Fairness Act that responds to the mandate to tackle unethical techniques and commercial practices related to dark-pattern marketing. The proposal converts the applicable directive on unfair commercial practices into a regulation and includes comprehensive rules and targeted simplifications, reducing burdens for businesses without a loss of consumer protection. The proposal is also coherent with the existing EU digital rulebook, including the Digital Services Act, Digital Markets Act, and AI Act, the GDPR, the Digital Omnibus and ongoing initiatives, such as the proposal for an EU KIDS Act.
More from supervisory authorities

New York workers data access rights: 8 November 2026, New York employees obtained new rights to access and challenge information in their personnel files, Ius Laboris law blog reports. Current and former employees may request a free copy of their personnel record up to twice per calendar year, and employers must produce it within five business days. Moreover, current employers must notify employees within 10 days of placing information in a personnel file that is, has been, or may be used to negatively affect the employee’s job standing. Violations are punishable by a fine of 500 to 2,500 dollars.
New York state had not previously granted a general statutory right for private-sector employees to inspect their own personnel files; historically, personnel records were treated as exclusively the employer’s property, the analysis states. These records can include anything from rate of pay and other compensation to performance evaluations, lists of probationary periods, and signed waivers.
Session replay: The French CNIL published its recommendation on Session Replay tools, which can be used for technical error detection, improving the user experience, and customer support, excluding ad retargeting. These tools record what a visitor does on a site or application (mouse, touch interactions, clicks, scrolling, entries in forms) to then replay their journey as a video. These tools work with trackers, which should be activated only if :
- the user’s consent is collected, for example, in your cookie banner (a consent management platform, or CMP);
- each purpose is presented before the visitor chooses;
- it is as simple to withdraw one’s consent as to give it;
- the collected data is deleted when no other legal basis justifies its retention.
The data controller must also be able to demonstrate at any time that consent has been obtained.
Email subject lines
Even when encryption technologies are used, email subject lines are not encrypted, as explained by the Liechtenstein data protection authority. This means they are always openly visible to anyone who intercepts and reads the email during transmission. For this reason, a subject line should never contain personally identifiable information. A pseudonymous case number or customer number may be used if it clearly does not represent sensitive personal data that could be misused for identity theft or fraud. This rule also applies if you are “only” replying to an email with a personal data subject line. Even then, the subject line should be adjusted.
Agentic AI security risks

Australia’s cybersecurity agency released the Securing AI testing environments paper to help organisations safely test AI models and agents used in cyber security activities. It explains how to establish secure, controlled testing environments that help contain AI agents and reduce the risk of unintended access to systems, networks and resources. The publication outlines key considerations for governance, monitoring, incident response planning, and human oversight during AI testing.
It also cautions against relying solely on prompts or built-in model safeguards to manage risk. Key considerations include:
- Conduct threat modelling to identify how AI agents could misuse tools, permissions or network access.
- Limit permissions and use short-lived credentials that can be quickly revoked when required.
- Test emergency shutdown procedures and ensure operators can isolate systems or stop testing when needed.
- Address vulnerabilities and use lessons learned from incidents and evaluations to improve future testing, etc.
Receive our digest by email
Sign up to receive our digest by email every 2 weeks
More enforcement decisions
Reason for dismissal: The Danish data protection authority Datatilsynet criticises an employer for sharing too much information about an employee’s dismissal in emails to all employees and selected partners. The employer did so with the desire to avoid rumours in the workplace and to protect the relationship with business partners. According to the employer, the former employee had contacted business partners and made accusations against the company. The authority, however, assessed that the employer’s interests did not outweigh the consideration for the former employee’s rights.
AI in elderly care: The Danish Datatilsynet also expressed serious criticism of a municipality’s inadequate risk assessment and lack of impact analysis when using AI in elderly care. A sensor unit in a nursing home used AI and camera-based technology to analyse movement patterns. The system provides insight into the residents’ sleep quality, restlessness, breathing, and movement patterns. The municipality has stated that the camera is only used in short sequences and only in the event of potential emergencies.
The authority says that the municipality should have conducted an impact assessment before the solution was put into use. The fact that the solution was only used for a trial period of six months does not change the municipality’s obligation to comply with data protection regulations.
It also pointed out that the rules on reassurance-creating welfare technology, under specific conditions, allow the use of the technology to ensure security, dignity and care for people with significantly and permanently reduced mental functions. The rules contain, among other things, special conditions for the use of AI, but the municipality had not included these rules in its considerations regarding the legal basis.
Meta multimillion fine over gallery pictures

The Italian Data Protection Authority fined Meta 805 million euros for using photos, videos and related metadata from Android users’ galleries without their specific consent. The feature must be disabled by default within 60 days, and consent must be requested separately. The fine concerns the “Get suggestions from your gallery when browsing Facebook” feature, available since 2023, which suggests photos and videos from the phone’s gallery as ideas to post while browsing.
The investigation revealed that:
- Access to the gallery is requested with a single permission, valid for the entire app and for multiple functions with different needs.
- The permission is preceded by a generic warning (“enable other functions”) that does not allow the user to understand that they are also authorising gallery suggestions.
- The feature was set to enabled by default. If granted in full, this type of permission allows the app to read the device’s entire photo archive, usually permanently, without the user or the authorities being able to verify which information is being used.
Location data in shopping apps
The Danish Datatilsynet investigated the Coop app and the Meny app and is seriously critical of the companies’ use of customers’ location information without a lawful basis for processing. Both have used the mobile phones’ built-in location data to identify which store customers were in when they used the scan-and-pay function.
The processing has been relatively limited. There has been no ongoing registration or tracking of customers’ movements, and only information about the relevant store has been used. However, the authority assesses that the use of mobile phone location data is an intrusive processing of personal data that requires special protection.
Both companies had referred to the rules on the performance of a contract and the balancing of interests. However, the regulator assesses that the processing was not necessary to offer the scan-and-pay function. The companies could have identified the relevant store in a less intrusive way, for example by having customers scan a QR code at the store’s entrance.
And Finally

The EDPB has picked the topic for its sixth coordinated enforcement action, which will concern automated decision-making (ADM), including AI-driven ADM, in recruitment. More specifically, the coordinated action will focus on the enforcement of the right not to be subject to decisions based solely on automated processing and the related right of access, in the context of the use of automated candidate screening tools, including AI-driven ones.
Data protection authorities across Europe will join this action voluntarily in the coming weeks, and the action itself will be launched in the course of 2027. Every annual action is designed to be flexible. DPAs can choose from a range of diverse organisations and actions in one or more of the following ways:
- fact-finding exercises,
- new formal investigations, and/or
- follow-up of ongoing formal investigations, etc.