What Happened?
The case involves a complainant who was part of an income protection insurance scheme organised by their employer. To claim benefits, the complainant had to attend medical appointments organised by an insurance company. However, without the complainant's knowledge, a third-party company shared information about these appointments, including details about the doctors involved, with the complainant's employer. The Data Protection Commission (DPC) found that the insurance company, as the data controller, breached data protection laws by sharing excessive information and failing to obtain explicit consent.
Because the data disclosed by the controller included specifics about the doctors' specialisations, it inferred the potential nature of the complainant's illness, thus warranting additional protection. The DPC affirmed that this extra safeguard imposed a prohibition on processing the data unless certain specified conditions were met, such as explicit consent from the complainant. The DPC deliberated on whether the complainant's signing of the claim form, which included consent for the controller to seek information, constituted explicit consent for disclosing information about medical appointments. While the complainant consented to the controller seeking such information, explicit consent for sharing it with third parties was not given. Consequently, the DPC concluded that the controller further breached legislation by sharing this data without explicit consent.
What does it mean for your organisation?
It is not enough to determine if the data controller has a legal ground to process special category of data, in particular health data in this case. It is relevantly important to take all necessary steps to protect data subject rights when their data is processed. Article 13 of the GDPR mandates that when personal data is collected from data subjects, the data controller must provide specific information, including the controller's identity, purpose of processing, recipients of the data, and data subject rights, to ensure fair and transparent data processing.
Article 14 applies when data is obtained from sources other than the data subject, requiring similar concise, transparent, and accessible information provision.
Furthermore, Article 5(1)(c) emphasises data minimisation, necessitating that personal data be relevant, limited, and stored for the shortest necessary period. Additionally, personal health data falls under the special category of personal data in Article 9 of the GDPR, subject to stricter regulations due to its sensitivity, requiring explicit consent for processing except in specific cases outlined in GDPR Article 9(2) and sections 45 to 54 of the Data Protection Act 2018.
Thus, when your organisation has a necessity to process special category of data, it is recommended to consider:
1) What exact information do you need and whether you can minimise its amount.
2) Do you have a lawful basis to process such information.
3) Do you inform data subject about all recipients of their personal data.
4) Do you have additional safeguard measures to protect special category of data.
5) Did you informed data subject of their rights when the data was collected.
All of those factors are equally important and should be properly noted in the relevant policies and procedures, that can be demonstrated not only to the data subject, but to the Data Protection Authority as well.
Tales from the coalface
Each month we look at the most prevalent Data Protection concerns raised by our clients. This month we were asked about AI Act and its interaction with the GDPR.
The EU Artificial Intelligence Act ("AI Act") was formally adopted by the European Parliament on March 13, 2024, with an overwhelming majority of votes in favour. This legislation represents a significant milestone as the world's inaugural comprehensive law specifically regulating AI, marking a historic moment for the European Union.
The EU legal framework for AI Act and the GDPR is complex and evolving, with some gaps and overlaps between different instruments. These legislative instruments have distinct scopes, definitions, and requirements, posing challenges for achieving compliance and maintaining consistency across regulatory landscapes. At the same time, the AI Act is akin to Europe’s General Data Protection Regulation (GDPR), but for artificial intelligence.
The GDPR regulates the management of personal data within the European Union (EU) and extends its reach extraterritorially, requiring compliance from companies outside the EU that process personal data of individuals in the EU. Similarly, the AI Act will have implications for companies around the world. While the GDPR focuses on the obligations of data controllers and processors, and the EU AI Act targets providers and users of AI systems, organisations will need to carefully align these frameworks to determine which entities are subject to the requirements of the GDPR, the AI Act, or both. This alignment is particularly important given the overlap between the two regulations, particularly in areas such as bias and discrimination, risk assessments, and solely automated decision-making.
Article 9 of the GDPR prohibits data controllers from processing special category data unless an exception applies. In contrast, the EU AI Act provides an explicit exemption from the Article 9 GDPR prohibition. Article 10(5) states that "to the extent that it is strictly necessary for the purposes of ensuring bias monitoring, detection and correction in relation to high-risk AI systems, the providers of such systems may process special categories of personal data referred to in Article 9(1) GDPR". Article 10(5) also mandates "appropriate" safeguards. Consequently, there is a lot of questions as to how those two requirements will coexist and will be used in practice.
Furthermore, Article 35 of the GDPR requires data controllers to conduct data protection impact assessments (DPIAs) when processing is likely to pose a high risk to the rights and freedoms of individuals. At the same time AI providers may not always be able to anticipate all potential uses of a system. An initial assessment by a provider to determine whether the system is high-risk according to the AI Act does not preclude a subsequent DPIA by the user, even if the provider's assessment concludes that the system is not high risk. Consequently, the same system could potentially be subject to different risk management requirements and classifications under each law.
Article 22(1) of the GDPR stipulates that individuals have the right not to be subject to decisions based solely on automated processing that yield legal or similarly significant outcomes. Conversely, Article 14 of the AI Act establishes a correlated necessity for human oversight of high-risk systems. Therefore, it appears conceivable, in principle, for an AI system classified as low or minimal risk under the AI Act to render a solely automated decision as defined by the GDPR without involving of human reconsideration.
Additionally, EU officials have indicated plans to develop more targeted AI laws post the June 2024 EU elections, potentially focusing on regulating AI use in employment. While the AI Act touches upon AI in employment, officials suggest further specific legislation could enhance this aspect.

