Navigating Privacy-Enhancing Technologies: Key Takeaways from the Inaugural Meeting of the Global PETs Network
In recent years, privacy-enhancing technologies (PETs) have been an increasingly popular subject on regulators’ and policymakers’ agendas. Whether by issuing guidance about these types of tools (Canada’s Office of the Privacy Commissioner; United Kingdom’s Information Commissioner’s Office; Organisation for Economic Co-operation and Development), setting up regulatory sandboxes (Singapore’s Personal Data Protection Commission; Colombia’s Superintendence of Industry and Commerce); or creating prize challenges (United States and United Kingdom),1 regulators are investing resources and energy to better understand, support the deployment, and potentially regulate PETs.
On June 26, 2023, the Israel Privacy Protection Authority (IPPA) and the Future of Privacy Forum (FPF) brought industry experts, government officials, and academia together in Tel Aviv to discuss experiences and challenges faced towards the adoption of PETs. The in-person event served as the inaugural meeting of an informal Global PETs Network for regulators, providing a platform to discuss the latest developments and projects related to privacy-enhancing technologies among regulators and relevant stakeholders worldwide.
The inaugural meeting, hereinafter referred to as the “PETs Conference,” included the presentation of two case studies, a closed roundtable for regulators, and an open discussion with academia and industry experts, with the discussions being held under Chatham House Rule. This blog analyzes the main challenges raised by participants for adequate implementation of privacy-enhancing technologies, as well as the main takeaways of the discussions.
PETs: an evolving concept gaining increasing attention
As technological developments increase the collection and exchange of personal data across jurisdictions and organizations, privacy-enhancing technologies can help by providing greater security, confidentiality, and protection of personal data. There are several types of PETs, which may be classified based on their functionality. For instance, some tools obfuscate and hide information (i.e., anonymization, synthetic data, differential privacy), other technologies allow for computations on encrypted data (i.e., homomorphic encryption, secure multi-party computation), while others facilitate the training of models without transferring and sharing data to a local server (i.e., federated learning).
Although PETs have received heightened attention from authorities in recent years through different policies and initiatives, the concept is not new. As a term of reference, PETs were first introduced by the predecessor of the Dutch Data Protection Authority and the Information and Privacy Commissioner in Ontario back in 1995, through a joint report that sought to demonstrate that identity-protective elements might be included in the design of information technology systems.2
Ever since, interest in PETs has increased not only through extensive research but also in practice. For instance, federated learning and multi-party computation have proven to be useful when feeding machine learning models with on-device user data to improve digital services and products, without transferring the data to a central server. Public and private sector players use differential privacy to protect identities and privacy of people when publishing large sets of data
While governments and organizations seem to acknowledge the potential benefits of PETs, significant challenges to their effective deployment remain. Some of these challenges include the lack of maturity and high costs associated with some of these technologies, as well as an apparent lack of communication between experts and regulators, resulting in limited regulatory guidance and understanding about the benefits, limitations, and use cases of PETs.
1. Collaboration and a greater understanding of PETs are essential
For some jurisdictions, privacy-enhancing technologies are still seen as a new and complex subject by regulators and companies alike. In that sense, educational resources and guidance can help translate the benefits and limitations of these tools. Although PETs may be encapsulated in one general concept, they differ in technical capabilities and usability. During the PETs Conference, participants praised efforts by regulators and international organizations to conceptualize the functionality and use cases of these technologies and tools. These studies include the United Kingdom’s Information Commissioner’s Office (ICO) recently published “PETs Guidance”, as well as the OECD’s 2023 Report on “Emerging Privacy-Enhancing Technologies.”
These efforts are a starting point for greater understanding and certainty about the role of PETs in protecting personal information and privacy, which can lead to more detailed guidance and initiatives. During the meeting, authorities advocated for increased communication and collaboration to advance the understanding of PETs. This not only refers to collaboration for new projects and initiatives but also to leveraging useful and already available information to provide greater guidance and certainty to the industry. For instance, authorities could translate and disseminate available guidance by foreign authorities to their own official languages, where useful, or build their own guidance from previous documents. This kind of exercise can be helpful in providing guidance to the industry faster and in building capacity and technical knowledge within agencies.
2. Regulatory certainty is necessary to boost the adoption of PETs
Secondly, regulatory guidance can provide greater certainty for the deployment and adoption of PETs. During the open discussion with industry stakeholders, some participants indicated that more certainty on regulators’ perceptions of these tools can spur the innovation and deployment of privacy-enhancing technologies. In the long term, guidance can also manifest through metrics to evaluate the success and risks associated with some of the tools.
While some organizations might recognize the value of these tools to offer more privacy-preserving products, barriers to their implementation – such as high investments in time and resources, technical expertise, lack of maturity, and information asymmetries between developers and potential buyers – are a major factor in deciding whether to invest in these tools. However, if these technologies provide an opportunity for more privacy-preserving products and services, regulators should make efforts to ensure that most organizations consider the implementation or integration of PETs, when possible. Authorities have an important role in building trust in the digital ecosystem by providing greater certainty regarding how privacy-enhancing technologies can ensure the protection of personal information.
Importantly, data protection authorities have a special task in identifying how PETs overlap with data protection principles and how these technologies could potentially complement data protection compliance systems. Providing greater certainty in this regard could be definitive to some organizations’ decisions regarding investment and adoption of privacy-enhancing technologies.
Later on in the open discussion, industry representatives highlighted the importance of noticing the dynamics and different incentives created by PETs. In providing guidance and regulatory certainty, authorities should consider that privacy-enhancing technologies can benefit different parties across the chain of data utilization, particularly in cases where a certain technique enables data-sharing across multiple organizations. In this sense, regulators could consider economic and behavioral incentives to foster collaboration between organizations and public institutions.
3. The adoption of PETs requires constant evaluation and review of potential detrimental market outcomes
Regulators and industry participants agreed on the merit of setting standards and certification programs as a viable way to generate more trust in the use and deployment of PETs across organizations. However, they also agreed that regulation and guidance are necessary to ensure the adequate implementation of standards. Importantly, regulators still have an important role in assessing whether standardized tools will be sufficient to comply with data protection regulations, and if additional measures are required to integrate data protection and privacy throughout organizations.
Finally, due to the high costs associated with some PETs, regulators should be cautious of potential barriers to competition that might arise from the deployment of these technologies. If PETs start to be actively promoted within digital services and products, centering privacy as a key market value, regulators must consider that certain companies might be able to get a competitive advantage through the early development and deployment of PETs. To avoid additional market deficiencies caused by privacy-enhancing technologies, regulators have an important task at hand in attempting to strike a balance between privacy and competition concerns.
Conclusion
Regulators, academia, and industry experts seem to agree that further study and understanding of the potential benefits and limitations of privacy-enhancing technologies is necessary. Importantly, if PETs are part of the solution towards privacy-enhanced products and services in the digital ecosystem, regulators must strengthen their efforts to achieve their adequate deployment. Particularly, data protection authorities must indicate the extent to which privacy-enhancing technologies align with data protection and privacy frameworks and should evaluate whether their implementation is enough, or if additional measures are necessary. This assessment requires greater communication and collaboration between regulators, academia, and industry.
Importantly, more regulatory certainty on whether and how organizations should deploy PETs is essential. PETs already face intrinsic barriers to their adoption because they require technical expertise within organizations and are costly to adopt. Regulatory certainty plays an important role in tackling these challenges by providing greater transparency and knowledge about PETs, as well as the technology’s relation to data protection compliance. Regulators and data protection authorities, in particular, should focus on providing more information about the potential of PETs and provide metrics to assess their effectiveness or risks, if possible.
Finally, while PETs can help build more privacy and trust in the digital ecosystem, it is important to note that they are not a fail-safe solution. Authorities and organizations should keep core data protection principles in mind and supplement these technical tools with other organizational and administrative measures.
1The US-UK PETs Prize Challenge was led by the U.K.’s Centre for Data Ethics and Innovation (CDEI) and Innovate UK, the U.S. National Institute of Standards and Technology (NIST), and the U.S. National Science Foundation (NSF), in cooperation with the White House Office of Science and Technology Policy.
EU’s Digital Services Act Just Became Applicable: Outlining Ten Key Areas of Interplay with the GDPR
DSA: What’s in a Name?
The European Union’s (EU) Digital Services Act (DSA) is a first-of-its-kind regulatory framework, with which the bloc hopes to set an international benchmark for regulating online intermediaries and improving online safety. The DSA establishes a range of legal obligations, from content removal requirements, prohibitions to engage in manipulative design and to display certain online advertising targeted to users profiled on the basis of sensitive characteristics, to sweeping accountability obligations requiring audits of algorithms and assessments of systemic risks for the largest of platforms.
The DSA is part of the EU’s effort to expand its digital regulatory framework to address the challenges posed by online services. It reflects the EU’s regulatory approach of comprehensive legal frameworks which strive to protect fundamental rights, including in digital environments. The DSA should not be read by itself: it is applicable on top of the EU’s General Data Protection Regulation (GDPR), alongside the Digital Markets Act (DMA), as well as other regulations and directives of the EU’s Data Strategy legislative package.
The Act introduces strong protections against both individual and systemic harms online, and also places digital platforms under a unique new transparency and accountability framework. To address the varying levels of risks and responsibilities associated with different types of digital services, the Act distinguishes online intermediaries depending on the type of business service, size, and impact, setting up different levels of obligations.
Given the structural and “systemic” significance of certain firms in the digital services ecosystem, the regulation places stricter obligations on Very Large Online Platforms (VLOPs) and Very Large Online Search Engines (VLOSEs). These firms will have to abide by higher transparency standards, provide access to (personal) data to competent authorities and researchers, and identify, analyze, assess, and mitigate systemic risks linked to their services. Such systemic risks have been classified into four different categories (Recitals 80-84): illegal content; fundamental rights (freedom of expression, media pluralism, children’s rights, consumer protection, and non-discrimination, inter alia); public security and electoral/democratic processes; and public health protection, with a specific focus on minors, physical and mental well-being, and gender-based violence.
The European Commission designated VLOPs and VLOSEs earlier this year (see Table 1), based on criteria laid out in the DSA and a threshold number of 45 million monthly users across the EU. The DSA obligations for these designated online platforms became applicable on August 25, 2023, with the exception of a transparency database whose publication was postponed for a month following complaints. The full regulation becomes applicable for all covered starting on February 17, 2024.
* The companies that sought to challenge their designation as ‘VLOPs’ – the European General Court will be addressing these challenges and will determine whether the European Commission’s designation shall be upheld.
However, VLOPs and VLOSEs are not the only regulated entities. All intermediaries that offer their services to users based in the EU, including online platforms such as app stores, collaborative economy platforms, and social media platforms, fall within the scope of the regulation, regardless of their number of users. Notably, micro and small-sized enterprises that do not meet the VLOP/VLOSE criteria, as defined by EU law, are exempted from some of the legal obligations. While “regular” online platforms may have scaled down requirements compared to VLOPs/VLOSEs, their new legal obligations are nonetheless significant and include, among others, transparency regarding their recommendation systems, setting up internal complaint-handling mechanisms, prohibitions on designing their platforms in a way that deceives or manipulates users, and prohibitions on presenting ads based on profiling using special categories of personal data, including personal data of minors.
All providers of intermediary services, including online platforms, covered by the DSA are also “controllers” under the GDPR to the extent that they process personal data and decide on the means and purposes of such processing. As a consequence, they have to comply with both these legal frameworks at the same time. While the DSA stipulates, pursuant to Recital 10, that the GDPR and the ePrivacy Directive serve as governing rules for personal data protection, some DSA provisions intertwine with GDPR obligations in complex ways, requiring further analysis. For instance, some of the key obligations in the DSA refer to “profiling” as defined by the GDPR, while others create a legal requirement for VLOPs and VLOSEs to give access to personal data to researchers or competent authorities.
After a brief overview of the scope of application of the DSA and a summary of its key obligations based on the type of covered entity (see Table 2), this blog maps out ten key areas where the DSA and the GDPR interact in consequential ways and reflects on the impact of this interaction on the enforcement of the DSA. The ten interplay areas we are highlighting are:
Manipulative design in online interfaces;
Targeted advertising based on sensitive data;
Targeted advertising and protection of minors;
Recommender systems free-of-profiling;
Recommender systems and advertising transparency;
Access to data for researchers and competent authorities;
Takedown of illegal content;
Risk Assessments;
Compliance function and the DSA legal representative;
Intermediary liability and the obligation to provide information.
The DSA Applies to Intermediary Services of Various Types and Sizes and Has Broad Extraterritorial Effect
The DSA puts in place a horizontal framework of layered responsibilities targeted at different types of online intermediary services, including:
Including “mere conduit services” (e.g. internet access, content delivery networks, WiFi hotspots); “caching services” (e.g. automatic, intermediate, and temporary storage of information); and “hosting services” (e.g. cloud and web-hosting services).
(2)Online platform services
Providers bringing together sellers and consumers, such as online marketplaces, app stores, collaborative economy platforms, social media platforms, and providers that disseminate information to the public.
(3) Very Large Online Platforms (VLOPs)
Reaching at least 45 million active recipients in the EU on a monthly basis (10% of the EU population).
(4)Very Large Online Search Engines (VLOSEs)
Reaching at least 45 million active recipients in the EU on a monthly basis (10% of the EU population).
Recitals 13 and 14 of the DSA highlight the importance of “disseminating information to the public” as a benchmark for which online platforms fall under the scope of the Regulation and the specific category of hosting services. For instance, Recital 14 explains that emails or private messaging services fall outside the definition of online platforms “as they are used for interpersonal communication between a finite number of persons determined by the sender of the communication.” However, the DSA obligations for online platforms may still apply to them if such services “allow the making available of information to a potentially unlimited number of recipients, … such as through public groups or open channels.”
Important carve-outs are made in the DSA for micro and small-sized enterprises, as defined by EU law, that do not meet the VLOP/VLOSE criteria. These firms are exempted from some of the legal obligations, in particular from making available an annual report on the content moderation they engage in, as well as the more substantial additional obligations imposed on providers of online platforms in Articles 20 to 28 – such as the prohibition to display ads based on profiling conducted on special categories of personal data, and obligations for platforms allowing consumers to conclude distance contracts with traders in Articles 29 to 32.
These carve-outs come in contrast with the broad applicability of the GDPR to entities of all sizes. This means, for instance, that even if micro and small-sized enterprises that are online platforms do not have to comply with the prohibitions related to displaying ads based on profiling using special categories of personal data and profiling of minors, they continue to fall under the scope of the GDPR and its requirements that impact such profiling.
The DSA has extra-territorial effect and global coverage, similar to the GDPR, since it captures companies regardless of whether they are established in the EU or not, as long as the recipients of their services have their place of establishment or are located in the EU (Article 2).
The DSA Just Became Applicable to VLOPs and VLOSEs and Will Continue to Roll Out to All Online Platforms
The Act requires that platforms and search engines publish their average monthly number of active users/recipients with the EU-27 (Article 24 – check the European Commission’s guidance on the matter). The first round of sharing those numbers was due on February 17, 2023. Based on the information shared through that exercise, the Commission designated the VLOPs and VLOSEs with additional obligations because of the “systemic risks that they pose to consumers and society,” (Article 33). The designation announcement was made public on April 25.
Four months after the designation, on August 25, 2023, the DSA provisions became applicable to VLOPs and VLOSEs through Article 92. This means that the designated platforms must already implement their obligations, such as conducting risk assessments, increasing transparency of recommender systems, and offering an alternative feed of content not subject to recommender systems based on profiling (see an overview of their obligations in Table 2).
As of February 17, 2024, all providers of intermediary services must comply with a set of general obligations (Articles 11-32), with certain exceptions for micro and small enterprises as explained above.
Table2 – List of DSA Obligations as Distributed Among Different Categories of Intermediary Service Providers
Pillar obligations
Set of Obligations
Intermediary Services
Hosting Services
Online Platforms
VLOPs/VLOSEs
Transparency measures
Transparency reporting (Article 15)
🚩
🚩
🚩
🚩
Requirements on terms and conditions wrt fundamental rights (Article 14)
🚩
🚩
🚩
🚩
Statement of reasons (Article 17)
🚩
🚩
Notice-and-action and obligation to provide information to users (Article 16)
🚩
🚩
🚩
Recommender system transparency (Articles 27 and 38)
🚩
🚩
User-facing transparency of online advertising (Article 24)
🚩
🚩
Online advertising transparency (Article 39)
🚩
User choice for access to information (Article 42)
🚩
Oversight structure to address the complexity of the online intermediary services ecosystem
Cooperation with national authorities following orders (Article 11)
🚩
🚩
🚩
🚩
Points of contact for recipients of service (Article 12) and, where necessary, legal representatives (Article 13)
🚩
🚩
🚩
🚩
Internal complaint and handling system (Article 20) and redress mechanism (Article 32) and out-of-court dispute settlement (Article 21)
🚩
🚩
Independent auditing and public accountability (Article 37)
🚩
Option for recommender systems not based on profiling (Article 38)
🚩
Supervisory fee (Article 43)
🚩
Crisis response mechanism and cooperation process (Article 36)
🚩
Manipulative Design
Online interface design and organization (Article 25)
🚩
🚩
Measures to counter illegal goods, services, or content online
Trusted flaggers (Article 22)
🚩
🚩
Measures and protection against misuse (Article 23)
🚩
🚩
Targeted advertising based on sensitive data (Article 26)
🚩
🚩
Online protection of minors (Article 28)
🚩
🚩
Traceability of traders (Articles 30-32)
🚩
🚩
Reporting criminal offenses (Article 18)
🚩
🚩
Risk management obligations and compliance officer (Article 41)
🚩
🚩
Risk assessment and mitigation of risks (Articles 34-35)
🚩
Codes of conduct (Articles 45-47)
🚩
Access to data for researchers
Data sharing with authorities and researchers (Article 40)
🚩
From Risk Assessments to Profiling and Transparency Requirements – Key Points of Interplay Between the DSA and GDPR
While the DSA and the GDPR serve different purposes and objectives at face value, ultimately both aim to protect fundamental rights in a data-driven economy and society, on the one hand, and reinforce the European single market, on the other hand. The DSA aims to establish rules for digital services and their responsibilities toward content moderation and combating systemic risks, so as to ensure user safety, safeguard fairness and trust in the digital environment, and enhance a “single market for digital services.” Notably, providing digital services is inextricably linked to processing data, including personal data. The GDPR seeks to protect individuals in relation to how their personal data is processed, ensuring that such processing respects their fundamental rights, while at the same time seeking to promote the free movement of personal data within the EU.
While the two regulations do not have the same taxonomy of regulated actors, the broad scope of the GDPR’s definitions of “controllers” and “processing of personal data” are such that all intermediaries covered by the DSA are also controllers under the GDPR in relation to any processing of personal data they engage in and for which they establish the means and purposes of processing. Some intermediaries might also be “processors” under the GDPR in specific situations, a fact that needs to be assessed on a case-by-case basis. Overall, this overlap triggers the application of both regulations, with the GDPR seemingly taking precedence over most of the DSA (Recital 10 of the DSA), with the exception of the intermediary liability rules in the DSA as the updated eCommerce Directive, which take precedence over the GDPR (Article 2(4) of the GDPR).
The DSA mentions the GDPR 19 times in its text across recitals and articles, with “profiling” as defined by the GDPR playing a prominent role in core obligations for all online platforms. These include the two prohibitions to display ads based on profiling that use sensitive personal data or the data of minors, and the obligation that any VLOPs and VLOSEs that use recommender systems must provide at least one option for their recommender systems not based on profiling. The GDPR plays an additional role in setting the definition for sensitive data (“special categories of data”) in its Article 9, which the DSA specifically refers to for the prohibition of displaying ads based on profiling done on such data. In addition to these cross-references, where it will be essential to apply the two legal frameworks consistently, there are other areas of overlap that create complexity for compliance, at the minimum, but also risks for inconsistencies (such as the DSA risk assessment processes and the GDPR Data Protection Impact Assessment). Additional overlaps may confuse individuals concerned regarding the best legal framework to rely on for removing their personal data from online platforms, as the DSA sets up a framework for takedown requests for illegal content that may also include personal data and the GDPR provides individuals with the right to obtain erasure of their personal data in specific contexts.
In this complex web of legal provisions, here are the elements of interaction between the two legal frameworks that stand out. As the applicability of the DSA rolls out on top of GDPR compliance programs and mechanisms, other such areas may surface.
Manipulative Design (or “Dark Patterns”) in Online Interfaces
These are practices that “materially distort or impair, either on purpose or in effect, the ability of recipients of the service to make autonomous and informed choices or decisions,” per Recital 67 DSA. Both the GDPR and the DSA address these practices, either directly or indirectly. The GDPR, on the one hand, offers protection against manipulative design in cases that involve processing of personal data. The protections are relevant for complying with provisions detailing lawful grounds for processing, requiring data minimization, setting out how valid consent can be obtained and withdrawn, or how controllers must apply Data Protection by Design and by Default when building their systems and processes.
Building on this ground, Article 25 of the DSA, read in conjunction with Recital 67, includes a ban on providers of online platforms to “design, organize or operate their online interfaces in a way that deceives or manipulates the recipients of their service or in a way that otherwise materially distorts or impairs the ability of the recipients of their service to make free and informed decisions.” The ban seems to be applicable only to online platforms as defined in Article 3(i) of the DSA, as a subcategory of the wide spectrum of intermediary services. Importantly, the DSA specifies that the ban on dark patterns does not apply to practices covered by the Unfair Commercial Practices Directive (UCPD) or the GDPR. Article 25(3) of the DSA highlights that the Commission is empowered to issue guidelines on how the ban on manipulative design applies to specific practices, so further clarity is expected. And since the protection vested by the GDPR against manipulative design will remain relevant and primarily applicable, it will be essential for consistency that these guidelines are developed in close collaboration with Data Protection Authorities (DPAs).
Targeted Advertising Based on Sensitive Data
Article 26(3) and Recital 68 of the DSA underline a prohibition of the providers of online platforms to “present” ads to users stemming from profiling them, as defined by Article 4(4) of the GDPR, based on sensitive personal data, as defined by Article 9 of the GDPR. Such personal data include race, religion, health status, and sexual orientation, among others on a limited list. However, it is important to mention that case law from the Court of Justice of the EU (CJEU) may further complicate the application of this provision. In particular, Case C-184/20 OT, in a judgment published a year ago, expanded “special categories of personal data” under the GDPR to also cover any personal data from which a sensitive characteristic may be inferred. Additionally, the very recent CJEU judgment in Case C-252/21 Meta v. Bundeskartellamtmakes important findings regarding how social media services as a category of online platforms can lawfully engage in profiling of their users pursuant to the GDPR, including for personalized ads. While the DSA prohibition is concerned with “presenting” ads based on profiling using sensitive data, rather than with the activity of profiling itself, it must be read in conjunction with the obligations in the GDPR for processing personal data for profiling and with the relevant CJEU case-law. To this end, the European Data Protection Board has published relevant guidelines for automated decision-making and profiling in general, but also specifically on targeting of social media users.
Targeted Advertising and Protection of Minors
Recital 71 of the GDPR already provides that solely automated decision-making, including profiling, with legal or similarly significant effects should not apply to children – a rule that is relevant for any type of context, such as educational services, and not only for online platforms. The DSA enhances this protection when it comes to online platforms, prohibiting the presentation of ads on their interface based on profiling by using personal data of users “when they are aware with reasonable certainty that the recipient of the service is a minor” (Article 28 of the DSA). Additionally, in line with the principle of data minimization provided by Article 5(1) of the GDPR, this DSA prohibition should not lead the provider of the online platform to “maintain, acquire or process” more personal data than it already has in order to assess if the recipient of the service is a minor. While this provision addresses all online platforms, VLOPs and VLOSEs are expected to take “targeted measures to protect the rights of the child, including age verification and parental control tools” as part of their obligation in Article 35(1)(j) to put in place mitigation measures tailored to their specific systemic risks identified following the risk assessment process. As highlighted in a recent FPF infographic and report on age assurance technology, age verification measures may require processing of additional personal data than what the functioning of the online service requires, which could be at odds with the data minimization principles in the absence of additional safeguards. This is an example where the two regulations complement each other.
In recent years, DPAs have been increasingly regulating the processing of personal data of minors. For instance, in the EU, the Irish Data Protection Commission published Fundamentals for a Child-Oriented Approach to Data Processing, the Italian Garante often includes the protection of children in its high-profile enforcement decisions (see, for instance, the TikTok and ChatGPT cases), and the CNIL in France published recommendations to enhance the protection of children online and launched several initiatives to enhance digital rights of children. This is another area where collaboration with DPAs will be very important for consistent application of the DSA.
Recommender Systems and Advertising Transparency
A significant area of overlap between the DSA and the GDPR relates to transparency. A key purpose of the DSA is to increase overall transparency related to online platforms, manifesting through several obligations, while transparency related to how one’s personal data are processed is an overarching principle of the GDPR. Relevant areas for this principle in the GDPR are found in Article 5, through extensive notice obligations in Articles 13 and 14, data access obligations in Article 15, and underpinned by modalities on how to communicate to individuals in Article 12. Two of the DSA obligations that increase transparency are laid out in Article 27, which imposes on providers of online platforms transparency related to how recommender systems work, and in Article 26, which imposes transparency related to advertising on online platforms. To implement the latter obligation, the DSA requires, per Recital 68, that the “recipients of a service should have information directly accessible from the online interface where the advertisement is presented, on the main parameters used for determining that a specific advertisement is presented to them, providing meaningful explanations of the logic used to that end, including when this is based on profiling.”
As for transparency related to recommender systems, Recital 70 of the DSA explains that online platforms should consistently ensure that users are appropriately informed about how recommender systems impact the way information is displayed and can influence how information is presented to them. “They should clearly present the parameters for such recommender systems in an easily comprehensible manner” to ensure that the users “understand how information is prioritized for them,” including where information is prioritized “based on profiling and their online behavior.” Notably, Articles 13(2)(f) and 14(2)(g) of the GDPR require that notices to individuals whose personal data is processed include “meaningful information about the logic involved, as well as the significance and the envisaged consequences” of automated decision-making, including profiling. These provisions should be read and applied together, complementing each other, to ensure consistency. This is another area where collaboration between DPAs and the enforcers of the DSA would be desirable. To understand the way in which DPAs have been applying this requirement so far, this case-law overview on automated decision-making under the GDPR published by the Future of Privacy Forum last year is helpful.
Recommender Systems Free-of-Profiling
“Profiling” as defined by the GDPR also plays an important role in one of the key obligations of VLOPs and VLOSEs: to offer users an alternative feed of content not based on profiling. Technically, this stems from an obligation in Article 38 of the DSA for VLOPs and VLOSEs to “provide at least one option for each of their recommender systems which is not based on profiling.” The DSA explains in Recital 70 that a core part of an online platform’s business is the manner in which information is prioritized and presented on its online interface to facilitate and optimize access to information for users: “This is done, for example, by algorithmically suggesting, ranking and prioritizing information, distinguishing through text or other visual representations, or otherwise curating information provided by recipients.”
The DSA text further explains that “such recommender systems can have a significant impact on the ability of recipients to retrieve and interact with information online, including to facilitate the search of relevant information,” as well as playing an important role “in the amplification of certain messages, the viral dissemination of information and the stimulation of online behavior.” Additionally, as part of their obligations to assess and mitigate risks on their platforms, VLOPs and VLOSEs may need to adjust the design of their recommender systems. Recital 94 of the DSA explains that they could achieve this “by taking measures to prevent or minimize biases that lead to the discrimination of persons in vulnerable situations, in particular where such adjustment is in accordance with Article 9 of the GDPR,” where Article 9 establishes conditions for processing sensitive personal data.
Access to Data for Researchers and Competent Authorities
Article 40 of the DSA includes an obligation for VLOPs and VLOSEs to provide access to the data necessary to monitor their compliance with the regulation to competent authorities (Digital Services Coordinators designated at the national level in the EU Member State of their establishment or the European Commission). This includes access to data related to algorithms, based on a reasoned request and within a reasonable period specified in the request. Additionally, they also have an obligation to provide access to vetted researchers following a request of their Digital Services Coordinator of establishment “for the sole purpose of conducting research that contributes to the detection, identification, and understanding of systemic risks” in the EU, and “to the assessment of the adequacy, efficiency, and impacts of the risk mitigations measures.” This obligation presupposes that the platforms may be required to explain the design, logic of the functioning, and the testing of their algorithmic systems, in accordance with Article 40 and its corresponding Recital 34.
Providing access to online platforms’ data entails, in virtually all cases, providing access to personal data as well, which brings this processing under the scope of the GDPR and triggers its obligations. Recital 98 of the DSA highlights that providers and researchers alike should pay particular attention to safeguarding the rights of individuals related to the processing of personal data granted by the GDPR. Recital 98 adds that “providers should anonymize or pseudonymize personal data except in those cases that would render impossible the research purpose pursued.” Notably, the data access obligations in the DSA are subject to further specification through delegated acts, to be adopted by the European Commission. These acts are expected to “lay down the specific conditions under which such sharing of data with researchers can take place” in compliance with the GDPR, as well as “relevant objective indicators, procedures and, where necessary, independent advisory mechanisms in support of sharing of data.” This is another area where the DPAs and the DSA enforcers should closely collaborate.
Takedown of Illegal Content
Core to the DSA are obligations for hosting services, including online platforms, to remove illegal content: Article 16 of the DSA outlines this obligation based on a notice-and-action mechanism initiated at the notification of any individual or entity. The GDPR confers rights on individuals to request erasure of their personal data (Article 17 of the GDPR) under certain conditions, as well as the right to request rectification of their data (Article 16 of the GDPR). These rights of the “data subject” under the GDPR aim to strengthen individuals’ control over how their personal data is collected, used, and disseminated. Article 3(h) of the DSA defines “illegal content” as “any information that, in itself or in relation to an activity … is not in compliance with Union law or the law of any Member State…, irrespective of the precise subject matter or nature of that law.” As a result, to the extent that “illegal content” as defined by the DSA is also personal data, an individual may potentially use either of the avenues, depending on how the overlap of the two provisions is further clarified in practice. Notably, one of the grounds for obtaining erasure of personal data is if “the personal data has been unlawfully processed,” and therefore processed not in compliance with the GDPR, which is Union law.
Article 16 of the DSA highlights an obligation for hosting services, including online platforms, to put mechanisms in place to facilitate the submission of sufficiently precise and adequately substantiated notices. Article 12 of the GDPR, on another hand, requires controllers to facilitate the exercise of data subject rights, including erasure, and to communicate information on the action taken without undue delay and in any case no longer than one month after receiving the request. The DSA does not prescribe a specific timeline to deal with notices for removal of illegal content, other than “without undue delay.” All hosting services and online platforms whose activity falls under the GDPR have internal processes set up to respond to data subject requests, which could potentially be leveraged in setting up mechanisms to remove illegal content pursuant notices as requested by the DSA. However, a key differentiator is that in the DSA content removal requests can also come from authorities (see Article 9 of the DSA) and from “trusted flaggers” (Article 22), in addition to any individual or entity – each of these situations under their own conditions. In contrast, erasure requests under the GDPR can only be submitted by data subjects (individuals whose personal data is processed), either directly or through intermediaries acting on their behalf. DPAs may also impose the erasure of personal data, but only as a measure pursuant to an enforcement action.
VLOPs/VLOSEs will have to additionally design mitigation measures ensuring the adoption of content moderation processes, including the speed and quality of processing notices related to specific types of illegal content and its expeditious removal.
Risk Assessments
The DSA, pursuant to Article 34, obliges VLOPs/VLOSEs to conduct a risk assessment at least once per year to identify, analyze, and assess “systemic risks stemming from the design or functioning of their service and its related systems,” including algorithmic systems. The same entities are very likely subject to the obligation to conduct a Data Protection Impact Assessment (DPIA) under Article 35 of the GDPR, as at least some of their processing operations, like using personal data for recommender systems or profiling users based on personal data to display online advertising, meet the criteria that trigger the DPIA obligation. A DPIA is required in particular where processing of personal data “using new technologies, and taking into account the nature, scope, context, and purposes of the processing, is likely to result in a high risk to the rights and freedoms of natural persons.”
There are four systemic risks that the DSA asks to be included in the risk assessment: dissemination of illegal content; any actual or foreseeable negative effects on the exercise of specific fundamental rights, among which the right to respect for private life and the right to the protection of personal data are mentioned; any actual or foreseeable negative effects on civic discourse, electoral processes and public security; and any actual foreseeable negative effects in relation to gender-based violence, the protection of public health and minors, and serious negative consequences to the person’s physical and mental well-being.
Among the elements that a DPIA under the GDPR must include is “an assessment of the risks to the rights and freedoms of data subjects” that may arise from how controllers process personal data through new technologies, such as algorithmic systems. Other elements that must be included are the measures envisaged to address these risks, similar to how Article 35 of the DSA requires VLOPs/VLOSEs to put mitigation measures in place tailored to the identified risks. The EDPB has also published guidelines on how to conduct DPIAs.
When conducting the risk assessments required by the DSA, VLOPs/VLOSEs must take into account whether and how specific factors enumerated in Article 34(2) influence any of the systemic risks mentioned. Most factors to consider are linked to how VLOPs/VLOSEs process personal data, such as the design of their algorithmic systems, the systems for selecting and presenting advertisements, and generally their data-related practices.
Both DSA risk assessments and DPIAs are ex-ante risk assessment obligations and both involve some level of engagement with supervisory authorities. The scope of the assessments differ, with the DSA focused on systemic risks and risks that go beyond impact on fundamental rights, and the GDPR’s DPIA focused on any risks that novel processing of personal data may pose on fundamental rights and freedoms and on assessments unique to data protection. However, they also have areas of clear overlap where processing of personal data is involved. DPIAs can potentially feed into DSA risk assessments, and the two processes should be implemented consistently.
Compliance Function and theDSA Legal Representative
Under the DSA, in accordance with Article 41, the designated VLOPs/VLOSEs will be obliged to establish a “compliance function,” which can be composed of several compliance officers. This function must be (i) independent from their operational functions; (ii) allocated with sufficient authority, stature and resources; and must have (iii) access to the management body of the provider to monitor the compliance of that provider with the DSA. On top of that, the compliance function will have to cooperate with the Digital Services Coordinator of the establishment, ensure that all risks are identified through the risk assessments and that the mitigation measures are effective, as well as inform and advise the management and employees of the provider in relation to DSA obligations.
All providers of the services designated as VLOPs and VLOSEs who are also controllers under the GDPR are under an obligation to appoint a Data Protection Officer (DPO), as they very likely meet the criteria required by Article 37 of the GDPR due to the nature and scope of their processing activities involving personal data. There are similarities between the compliance function and the DPO, including their independence, reporting to the highest management level, their key task to monitor compliance with the whole regulation that creates their role, or their task to cooperate with the competent supervisory authorities. Appointing two independent roles that have a powerful internal position and with roles that may overlap to a certain extent will require consistency and coordination, which can be supported by further guidance from DPAs and DSA supervisory authorities.
Another role in the application of the two regulations that has many similarities is the role of a “representative” in the EU, in the situations of extraterritorial applicability of the DSA and the GDPR covering entities that do not have an establishment in the EU. In the DSA, this obligation pertains to all online service providers, pursuant to Article 13. If they are processing personal data in the context of targeting their services to individual recipients in the EU or if they monitor the recipients’ behavior, the service provider triggers the extraterritorial application of the GDPR as well. In such cases, they also need to appoint a GDPR representative, in accordance with Article 27. Under the GDPR, the representative acts as a mere “postal box” or point of correspondence between the non-EU controller and processor on one hand and DPAs or data subjects on the other hand, with liability that does not go beyond its own statutory obligations. In contrast, Article 13(3) of the DSA suggests that the “legal representative” could be held liable for failures of the intermediary service providers to comply with the DSA. Providers must mandate their legal representatives for the purpose of being addressed “in addition to or instead of” them by competent authorities, per Article 13(2) of the DSA.
Recital 44 of the DSA clarifies that the obligation to appoint a “sufficiently mandated” legal representative “should allow for the effective oversight and, where necessary, enforcement of this regulation in relation to those providers.” The legal representative must have “the necessary powers and resources to cooperate with the relevant authorities” and the DSA envisages that there may be situations where providers even appoint in this role “a subsidiary undertaking of the same group as the provider, or its parent undertaking, if that subsidiary or parent undertaking is established in the Union.” Recital 44 of the DSA also clarifies that the legal representative may also only function as a point of contact, “provided the relevant requirements of this regulation are complied with.” This could mean that if other structures are in place to ensure an entity on behalf of the provider can be held liable for non-compliance by a provider with the DSA, the representative can also function just as a “postal box.”
Intermediary Liability and the Obligation to Provide Information
Finally, the GDPR and the DSA intersect in areas where data protection, privacy, and intermediary liability overlap.
The GDPR, per Article 2, stresses that its provisions shall be read without prejudice to the e-Commerce Directive (2000/31/EC), in particular, to “the liability rules of intermediary service providers in Articles 12 to 15 of that Directive”. However, the DSA, pursuant to its Article 89, stipulates that while Articles 12 to 15 of the e-Commerce Directive become null, relevant “references to Articles 12 to 15 of Directive 2000/31/EC shall be construed as references to Articles 4, 5, 6 and 8 of this Regulation, respectively.”
The DSA deals with the liability of intermediary services providers, especially through Articles 4 to 10. With respect to Article 10, addressing orders to provide information, the DSA emphasizes the strong envisaged cooperation between intermediary service providers, national authorities, and also the Digital Services Coordinators as enforcers. This could potentially involve the sharing of information, including in certain cases that already collected personal data, in order to combat illegal content online. The GDPR actively passes the baton on intermediary liability on the DSA, but in the eventuality of data sharing and processing, the intermediary service providers should ensure that they comply with the protections of the GDPR (in particular sections 2 and 3). This overlap signals yet another instance where the two Regulations will be complementary to each other, this time in the case of intermediary liability and the obligation to provide information.
The DSA Will Be Enforced Through a Complex Web of Authorities, And The Interplay With The GDPR Complicates It
Enforcement in such a complex space will be challenging. In a departure from the approach promoted by the GDPR, where enforcement is ensured primarily at the national level and through the One Stop Shop mechanism for cross-border cases coordinated through the European Data Protection Board, the DSA centralizes enforcement of the DSA at the EU level when it comes to VLOPs and VLOSEs, leaving it in the hands of the European Commission. However, Member States will also be playing a role in ensuring enforcement of the DSA against the intermediary services providers who are not VLOPs and VLOSEs. Each Member State must designate one or more competent authorities for the enforcement of the DSA, and if they designate more, they must choose one to be appointed as their Digital Services Coordinator (DSC). The deadline to designate DSCs is February 2024. Challenges come with the designation of national competent authorities left to the Member States, as it seems that there is no consistent approach related to what type of authority will be most appropriately positioned to enforce the Act. Not all Member States have appointed their DSCs for the time being, but there is a broad spectrum of enforcers that Member States plan to rely on, creating a scattered landscape.
Table 3 – Authorities Designated or Considered for Designation as Digital Services Coordinators Across the EU Member States (Source: Euractiv)
Digital Services Coordinators
Member States
Media Regulator
Belgium, Hungary, Ireland and Slovakia
Consumer Protection Authority
Finland and the Netherlands
Telecoms Regulator
Czech Republic, Germany, Greece, Italy, Poland, Slovenia and Sweden
Competition Authority
Spain
The Digital Services Coordinators will be closely collaborating and coordinating with the European Board for Digital Services, which will be undertaking an advisory capacity (Articles 61-63 of the DSA), in order to ensure consistent cross-border enforcement. Member States are also tasked to adopt national rules on penalties applicable to infringements of the DSA, including fines that can go up to 6% of the annual worldwide turnover of the provider of intermediary services concerned in the preceding financial year (Article 52 of the DSA). Complaints can be submitted to DSCs by recipients of the services and by any body, organization, or association mandated to exercise rights conferred by the DSA to recipients. With respect to VLOPs and VLOSEs, the European Commission can issue fines not exceeding 6% of the annual worldwide turnover in the preceding year, following decisions of non-compliance which can also ask platforms to take necessary measures to remediate the infringements. Moreover, the Commission can also order interim measures before an investigation is completed, where there is an urgency due to the risk of serious damage to the recipients of the service.
The recipients of the service, including users of online platforms, also have a right to seek compensation from providers of intermediary services for damages or loss they suffer due to infringements of the DSA (Article 54 of the DSA). The DSA also applies in out-of-court dispute resolution mechanisms with regard to decisions of online platforms related to illegal content (Article 21 of the DSA), independent audits in relation to how VLOPs/VLOSEs comply with their obligations (Article 37 of DSA), and voluntary codes of conduct adopted at the Union level to tackle various systemic risks (Article 45), including codes of conduct for online advertising (Article 46) and for accessibility to online services (Article 47).
The newly established European Centre for Algorithmic Transparency (ECAT) also plays a role in this enforcement equation. The ECAT will be supporting the Commission in its assessment of VLOPs/VLOSEs with regard to risk management and mitigation obligations. Moreover, it will be particularly relevant to issues pertaining to recommender systems, information retrieval, and search engines. The ECAT will use a principles-based approach to assessing fairness, accountability, and transparency. However, the DSA is not the only regulation relevant to the use of algorithms and AI by platforms: the GDPR, the upcoming Digital Markets Act, the EU AI Act, and the European Data Act add to this complicated landscape.
The various areas of interplay between the DSA and the GDPR outlined above require consistent interpretation and application of the law. However, there is no formal role recognized in the enforcement and oversight structure of the DSA for cooperation or coordination, specifically among DPAs, the European Data Protection Board, or the European Data Protection Supervisor. This should not be an impediment to setting up processes for such cooperation and coordination within their respective competencies, as the rollout of the DSA will likely reveal the complexity of the interplay between the two legislative frameworks even beyond the ten areas outlined above.
Editor: Alexander Thompson
FPF Submits Comments to the FTC on the Application for a New Parental Consent Method
Today, the Future of Privacy Forum (FPF) submitted comments to the Federal Trade Commission (FTC) regarding the use of “Privacy-Protective Facial Age Estimation” as a potential mechanism for verifiable parental consent (VPC) under the Children’s Online Privacy Protection Act (COPPA) Rule.
FPF observes:
The “Privacy-Protective Facial Age Estimation” technology may improve the existing landscape for verifiable parental consent, provided appropriate privacy safeguards are in place;
The “Privacy-Protective Facial Age Estimation” technology and associated risks are distinct from the biometric privacy risks associated with facial recognition technologies; and
If the FTC approves the application, the Commission’s approval should require ongoing implementation of the privacy and fairness safeguards outlined in the application.
In June, FPF published The State of Play: Is Verifiable Parental Consent Fit for Purpose?, investigating the shortcomings and opportunities presented by the current framework for verifiable parental consent (VPC) under COPPA and encouraging ingenuity to address key challenges. As federal lawmakers seek more comprehensive ways to update the 1998 law to match the 2023 online landscape, the approval of a new method for obtaining VPC has the potential to improve a process that is grappling with changing technologies, business practices, and individuals’ expectations.
FPF’s comments do not discuss the merits of using technology as a method of age estimation or verification for all users of a child-directed or mixed-audience service, which may place disproportionate privacy risks and burden on all users. Rather, we confine our analysis to the proposed context of this application, which we understand to only refer to the limited use of verifying that a purported parent granting COPPA consent is, in fact, an adult.
FPF’s full comments to the Commission are available here.
Data Sharing for Research: A Compendium of Case Studies, Analysis, and Recommendations
Today, the Future of Privacy Forum (FPF) published a report on corporate-academic partnerships that provides practical recommendations for companies and researchers who want to share data for research. The Report, Data Sharing for Research: A Compendium of Case Studies, Analysis, and Recommendations, demonstrates how, for many organizations, data-sharing partnerships are transitioning from being considered an experimental business activity to an expected business competency.
Corporate data-sharing partnerships offer compelling benefits to companies, researchers, and society to drive progress in a broad array of fields. However, organizations have long faced complex commercial, legal, ethical, and reputational risks that accompany the activity and act as disincentives to sharing data for academic research.
This report contains eight case studies that look at specific corporate/academic data-sharing partnerships in depth, from initiation through the publication of research findings. These case studies illuminate practical challenges for implementing corporate data sharing with researchers. Some common themes that emerged from the case studies include:
Successful data-sharing partnerships use Data-Sharing Agreements that require both the company and researchers to take steps to protect privacy.
Some of the challenges of data sharing include technical knowledge and infrastructure gaps between companies and researchers, and the continuing need for ethics and privacy review for industry-based research.
Promising practices for data sharing include the use of Privacy Enhancing Technologies and company-created, public-facing data-sharing menus to facilitate new partnerships.
While data sharing has significant costs and inherent risks, the risks can be managed, and the benefits to researchers, companies, and society make data sharing worth the effort.
This report builds upon prior FPF research, including the publication of The Playbook: Data Sharing for Research and the companion infographic in 2022. The case studies examine how data sharing works in a practical environment. By analyzing the case studies as a group, we arrived at recommendations for all parties interested in pursuing an ethical data-sharing partnership that protects against privacy risks.
For companies considering data sharing for research, we recommend the following:
Create a public webpage listing all data the company is willing to share, describe any requirements for potential data-sharing partnerships, and create a public form for researchers to ask questions.
Bolster privacy by using Privacy Enhancing Technologies (PETs), reduce data sensitivity through data minimization and aggregation, and include metadata as part of internal privacy reviews before sharing.
Promote rigorous data governance by assigning multiple people with expertise to manage data sharing, connect core team members to the data-sharing team, and adapt Data Sharing Agreements to align with the company’s available budgetary and personnel support.
Ensure researchers maintain authorial control over research methods, data analysis, interpretation, and publishing/communication venue. Where appropriate, companies may reserve the right to review data before publication to assess privacy risks and consult on the analytical limitations of the data.
For researchers interested in using data held by a company for research, we recommend the following:
Proactively contact companies that may hold data of interest and maintain continuous communication, especially about publication expectations.
Cultivate internal partnerships by involving the university general counsel early on and checking to see if the university has a standard Data Sharing Agreement. Contact the university’s Research Integrity Office and Information Technology Office before any data is shared, and consult the library for research support.
Receive training on how to integrate Privacy Enhancing Technologies in research and include privacy-related technical infrastructure in all funding proposals.
Coordinate with the company about any requirements for publishing, data sharing, data retention, and citation while maintaining academic independence.
You can access each of our individual case studies at these links:
Download accessible versions of these documents here.
FPF offers the Ethics and Data in Research Working Group, which analyzes US legislation impacting research and data, discusses ethical and technological research challenges, and develops best practices for privacy protection, risk reduction, and data sharing in research. Learn more and join the Working Group here.
For inquiries about this report, please contact Shea Swauger, Senior Researcher for Data Sharing and Ethics, at [email protected].
This project is supported by the Alfred P. Sloan Foundation, a not-for-profit grantmaking institution whose mission is to enhance the welfare of all through the advancement of scientific knowledge.
The Digital Personal Data Protection Act of India, Explained
Authors: Raktima Roy, Gabriela Zanfir-Fortuna
Raktima Roy is a Privacy Attorney with several years of experience in Indiaand holds an LLM in Law and Technology from Georgetown University, as well as an FPF Global Privacy Intern.
The Digital Personal Data Protection Act of India (DPDP) sprinted through its final stages last week after several years of debates, postponements and negotiations, culminating with its publication in the Official Gazette on Friday, August 11, 2023. In just over a week, the Bill passed the lower and upper Houses of the Parliament and received Presidential assent. India, the most populous country in the world with more than 1.4 billion people, is the largest democracy and the 19th country among the G20 members to pass a comprehensive personal data protection law – which it did during its tenure holding the G20 Presidency.
The adoption of the DPDP Bill in the Parliament comes 6 years after Justice K.S. Puttaswamy v Union of India, a landmark case in which the Supreme Court of India recognized a fundamental right to privacy in India, including informational privacy, within the “right to life” provision of India’s Constitution. In this judgment, a nine-judge bench of the Supreme Court urged the Indian Government to put in place “a carefully structured regime” for the protection of personal data. As part of India’s ongoing efforts to create this regime, there have been several rounds of expert consultations and reports, and two previous versions of the bill were introduced in the Parliament in 2019 and 2022. A brief history of the law is available here.
The law as enacted is transformational. It has a broad scope of application, borrowing from the EU’s General Data Protection Regulation (GDPR) approach when defining “personal data” and extending coverage to all entities who process personal data regardless of size or private status. The law also has significant extraterritorial application. The DPDP creates far reaching obligations, imposing narrowly defined lawful grounds for processing any personal data in a digital format, establishing purpose limitation obligations and their corollary – a duty to erase the data once the purpose is met, with seemingly no room left for secondary uses of personal data, and creates a set of rights for individuals whose personal data are collected and used, including rights to notice, access and erasure. The law also creates a supervisory authority, the Data Protection Board of India (Board), which has the power to investigate complaints and issue fines, but does not have the power to issue guidance or regulations.
At the same time, the law provides significant exceptions for the central government and other government bodies, the degree of exemption depending on their function (such as law enforcement). Other exemptions include those for most publicly available personal data, processing for research and statistical purposes, and processing the personal data of foreigners by companies in India pursuant a contract with a foreign company (such as outsourcing companies). Some processing by startups may also be exempt, if notified by the government. The Act also empowers the central government to act upon a notification by the Board and request access to any information from an entity processing personal data, an intermediary (as defined by the Information Technology Act, 2000 – the “IT Act”) or from the Board, as well as to order suspension of access of the public to specific information. The Central Government is also empowered to adopt a multitude of “rules” (similar to regulations under US state privacy laws) that detail the application of the law.
It is important to note that the law will not come into effect until the government provides notice of an effective date. The DPDP Act does not contain a mandated transitional period akin to the two-year gap between the 2016 enactment of the GDPR and its entry into force in May 2018. Rather, it empowers the Government to determine the dates on which different sections of the Act will come into force, including the sections governing the formation of the new Board that will oversee compliance with the law.
This blog will lay out the most important aspects of the DPDP Act, understanding nonetheless that many of its key provisions will be shaped up through subsequent rules issued by the central government, and through practice.
The DPDP Act Applies to “Data Fiduciaries,” “Significant Data Fiduciaries,” and provides rights for “Data Principals”
The DPDP Act seeks to establish a comprehensive national framework for processing personal data, replacing a much more limited data protection framework under the IT Actand rules that currently provide basic protections to limited categories of “sensitive” personal data such as sexual orientation, health data, etc. The new law by contrast covers all “personal data” (defined as “any data about an individual who is identifiable by or in relation to such data”) and does not contain heightened protection for any special category of data. The definition of “personal data,” thus, relies on the broad “identifiability” criterion, similar to the GDPR. Only “digital” personal data, or personal data collected through non-digital means that have been digitized subsequently are covered by the law.
The DPDP Act uses the term “data principal” to refer to the individual that the personal data relates to (the equivalent of “data subject” under the GDPR). A “data fiduciary” is the entity that determines the purposes and means of processing of personal data, alone or in conjunction with others, and is the equivalent to a “data controller” under GDPR. While the definition of data fiduciaries includes a reference to potential joint fiduciaries, the Act does not provide any other details about this relationship.
The definition of fiduciaries does not distinguish between private and public, natural and legal persons, technically extending to any person as long as the other conditions of the law are met.
Specific Fiduciaries, Public or Private, Are Exempted or May Be Exempted from the Core Obligations of the Act
The law includes some broad exceptions for government entities in general, and others apply to specific processing purposes. For instance, the law allows the government to exempt activities that are in the interests of the sovereignty and integrity of India, the security of the State, friendly relations with foreign States, maintenance of public order, or preventing incitement to commit crimes if it provides notice of the exemptions. Justice Srikrishna, who as the head of an expert committee set up to recommend a data protection law in India led the creation of the 2017 first draft of the law, has been critical of these government exemptions, as have been several Members of Parliament during the legislative debate.
Some targeted exceptions also apply to companies, and are either well defined in the law or left to the government for specification. Under what can be called an “outsourcing exception,” the Act exempts companies based in India who process the personal data of people outside of India pursuant to a contract with a company based outside of India from core DPDP obligations including the rights of access and erasure normally held by data principals. Instead, such companies are largely required to only comply with data security obligations.
In addition, the government is empowered to exempt any category of data fiduciaries from some or all of the law, with the DPDP itself referring to “startups” in this context. These are fairly broad provisions and do not include any guidance on how they will apply or who could benefit from them. The government will need to make a specific designation for this exception to operate.
Significant Data Fiduciaries Have Significant New Obligations, such as DPOs, DPIAs and Audits
The DPDP Act empowers the Government to designate any data fiduciary or class of data fiduciaries as a “Significant Data Fiduciary” (SDF), which is done using a series of criteria that lack quantifiable thresholds. These factors range from assessing characteristics of the processing operations (volume and sensitivity of personal data processed and the risk posed to the rights of data principals), to broader societal and even national sovereignty concerns (potential impact of the processing on the sovereignty and integrity of India; risk to electoral democracy; security of the state; and public order).
The designation of companies as SDFs is consequential, because it comes with enhanced obligations. Chief among them, SDFs will need to appoint a Data Protection Officer (DPO), who must be based in India and be the point of contact for a required grievance redressal mechanism. SDFs must also appoint an independent data auditor to carry out data audits and evaluate the SDF’s compliance with the DPDP Act, and to undertake periodic Data Protection Impact Assessments.
It is important to note that appointing a DPO is not an obligation for all data fiduciaries. However, all fiduciaries are under an obligation to establish a “readily available” mechanism for redressing grievances by data principals in a timely manner. In order for such a process to be operationalized, usually an internal privacy compliance function or a dedicated privacy officer would be helpful.
The DPDP Act Recognizes the Role of Data Processors
Data processors are recognized by the DPDP Act, which makes it clear that fiduciaries may engage, appoint or otherwise involve processors to process personal data on their behalf “only under a valid contract” (Section 8(2)). There are no prescribed rules for what a processing contract should entail. However, the DPDP Act places all obligations on data fiduciaries, which remain liable for complying with the law.
Data fiduciaries remain liable for overall compliance, regardless of any contractual arrangement to the contrary with data processors. The DPDP Bill requires data fiduciaries to mandate that a processor delete data when a data principal withdraws consent, and fiduciaries be able to share information of processors they have engaged when requested by a data subject.
The DPDP Act Has Broad Extraterritorial Effect and Almost No Restrictions for International Data Transfers
The DPDP Act applies to the processing of “digital personal data” within India. Importantly, the definition of the “data principal” does not include any condition related to residence or citizenship, meaning that it is conceivable fiduciaries based in India who process the personal data of foreigners within the territory of the country may be covered by the Act (outside of the “outsourcing exception” mentioned above).
The Act also applies extraterritorially to processing of digital personal data outside India, if such processing is in connection with any activity related to offering of goods or services to data principals within India. The extraterritorial effect is similar in scope to the GDPR, and it may leave room for a broader interpretation through its inclusion of “any activity” connected to the offering of goods or services.
The DPDP Act does not currently restrict the transfer of personal data outside of India. It reverses the typical paradigm of international data transfer provisions in laws like the GDPR, by presuming that transfers may occur without restrictions, unless the Government specifically restricts transfers to certain countries (blacklisting) or enacts any other form of restriction (Section 16). No criteria for such restrictions have been mentioned in the law. This is a significant departure from previous instances of the Bill, which at one point contained data localization obligations (2018), and evolved at another point into “whitelisting” of countries (2022).
It should also be noted that other existing sectoral laws (e.g., those governing specific industries like banking and telecommunications) already contain restrictions on cross-border transfers of particular kinds of data. The DPDP Act clarifies that existing localization mandates will not be affected by the new law.
Consent Remains Primary Means for Lawful Processing of Personal Data Under the Act
Data fiduciaries are under an obligation to process personal data for a lawful purpose and only if they either obtain consent from the data principal for that purpose, or they identify a “legitimate use” consistent with Section 4. This process is conceptually similar to the approach proposed by the GDPR, requiring a lawful ground before personal data can be collected or otherwise processed. However, in contrast to the GDPR (which provides for six possible lawful grounds), the DPDP Act includes only two: strictly defined “consent” and “legitimate use.”
Which lawful ground is used for a processing operation is consequential. Based on the wording of the Act and in the absence of further specification, the obligations of fiduciaries to give notice and respond to access, correction and erasure requests (see Section 4 of this blog) are only applicable if the processing is based on consent and on voluntary sharing of personal data by the principal.
Valid Consent Has Strict Requirements, Is Withdrawable, And Can be Exercised Through Consent Managers
The DPDP Act requires that consent for processing of personal data be “free, specific, informed, unconditional and unambiguous with a clear affirmative action.” These conditions are similarly strict to those required under the GDPR, highlighting that the people whose personal data are processed must be free to give consent, and their consent must not be tied to other conditions.
In order to meet the “informed” criterion, the Act requires that notice be given to principals before or at the time that they are asked to give consent. The notice must include information about the personal data to be collected, the purpose for which it will be processed, the manner in which data principals may exercise their rights under the DPDP Act, and how to make a complaint to the Board. Data principals must be given the option to receive the information in English or a local language among the languages specified in the Constitution.
The DPDP Act addresses the issue of legacy data for which companies may have received consent prior to the enactment of the law. Fiduciaries should provide the same notice to these data principals as soon as “reasonably practicable.” In that case, however, the data processing may continue until the data principal withdraws consent.
Data fiduciaries may only process personal data for the specific purpose provided to the data principal and must obtain separate consent to process the data for a new purpose. In practice, this will make it difficult for data fiduciaries to rely on “bundled consent.” Provisions around “secondary uses” of personal data or “compatible purposes” are not addressed in the Act, making the purpose limitation requirements strict.
Data principals may also withdraw their consent at any time – and data fiduciaries must ensure that the process for withdrawing consent is as straightforward as that for giving consent. Once consent is withdrawn, personal data must be deleted unless a legal obligation to retain data applies. Additionally, data fiduciaries must ask any processors to cease processing any personal data for which consent has been withdrawn, in the absence of legal obligations imposing data retention.
The DPDP Act allows principals to give, manage, review and withdraw their consent through a “Consent Manager,” which will be registered with the Board and must provide an accessible, transparent, and interoperable platform. Consent Managers are part of India’s “Data Empowerment And Protection Architecture” policy, and similar structures have been already functional for some time, such as in the financial sector. Under the DPDP Act, Consent Managers will be accountable to data principals and act on their behalf as per prescribed rules. The Government will notify (in the Gazette) the conditions necessary for a company to register as a Consent Manager, which may include fulfilling minimum technical or financial criteria.
“Legitimate Uses” Are Narrowly Defined and Do Not Include Legitimate Interests or Contractual Necessity
As alternative to consent, all other lawful grounds for processing personal data have been amalgamated under the “legitimate uses” section, including some grounds of processing that previously appeared under a “reasonable purposes” category in previous iterations of the bill. It is notable that the list of “legitimate uses” in Section 7 of the Act does not include similar provisions to the grounds of “contractual necessity” and “legitimate interests” found in GDPR-style data protection laws, leaving limited options to private fiduciaries for grounding processing of personal data outside of consent, including for routine or necessary processing operations.
Among the defined “legitimate uses”, the most relevant ones for processing personal data outside of a government, emergency or public health context, are the “voluntary sharing” of personal data under Section 7(a) and the “employment purposes” use under Section 7(i).
The lawful ground most likely to raise interpretation questions is “voluntary sharing.” It allows a fiduciary to process personal data for a specified purpose for which a principal has voluntarily provided their personal data to the data fiduciary (presumably, provided it without the fiduciary seeking to obtain consent), and for which the principal has not indicated to the fiduciary an objection to the use of the personal data. For instance, one of the illustrations included in the law to explain Section 7(a) is the hypothetical of a buyer requesting a receipt of purchase at a store be sent to her phone number, permitting the store to use the number for that purpose. There is a possibility that subsequent rules may expand this “legitimate use” to cover instances of “contractual necessity” or “legitimate interests.”
A fiduciary may also process personal data without consent for purposes of employment or those related to safeguarding the employer from loss or liability, such as prevention of corporate espionage, maintenance of confidentiality of trade secrets, intellectual property, classified information or provision of any service to employees.
Data Principals Have a Limited Set of “Data Subject Rights,” But Also Obligations
The DPDP Act provides data principles a set of enumerated rights, which is limited compared to those offered under modern GDPR-style data protection laws. The DPDP guarantees a right of access and a right to erasure and correction, in addition to a right to receive notice before consent is sought (similar to the right to information in the GDPR). Thus, a right to data portability, a right to object to processing based on other grounds than consent, and the right not to be subject to solely automated decision-making are missing.
Instead, the DPDP Act provides for two other rights – a right to “grievance redressal,” which entails the right to have an easily accessible point of contact provided by the fiduciary to respond to complaints from the principal, and a right to “appoint a nominee,” which permits the data principal to nominate someone who can exercise rights on their behalf in the event of death or incapacity.
Notably, the rights of access, erasure and correction are limited to personal data processing based on consent or the “voluntary disclosure,” legitimate use, which means that whenever government bodies or other fiduciaries rely on any of the “legitimate uses” grounds they will not need to reply to access or erasure/correction requests, unless further rules adopted by the government specify otherwise.
In addition, the right of access is quite limited in scope. It only gives data principals the right to request and obtain a summary of the personal data being processed and of the relevant processing activities (as opposed to obtaining a copy of the personal data), and the identities of all fiduciaries and processors with whom the personal data has been shared by the fiduciary, along with a summary of the data being shared. However, Section 11 of the law leaves space for subsequent rules that may specify additional information to be given access to.
Data principals have the right to request erasure of personal data pursuant to Section 12(3), but it is important to highlight that erasure may also be required automatically – after the withdrawal of consent or when the specified purpose is no longer being served (Section 8(7)(a)). Similarly, correction, completion and updating of personal data can be requested by the principal, but must also occur automatically when the personal data is “likely to be used to make a decision that affects” the principal (Section 8(3)).
Data Principals May Be Fined if They Do Not Comply With Their Obligations
Unlike the majority of international data protection laws, Section 15 of the DPDP Act imposes duties on data principals, similar to Article 10 of Vietnam’s recently adopted Personal Data Protection Decree (titled “Obligations of data subjects”).
These obligations include, among others, a duty not to impersonate someone else while providing personal data for a specified purpose, not suppress any material information while providing personal data for any document issued by the Government, and, significantly, not register a false or frivolous grievance or complaint. Noncompliance may result in a fine (see clause 5 of the Schedule). This may hamper the submission of complaints with the Board, per expert analysis.
Fiduciaries are Bound by a Principle of Accountability and Have Data Breach Notification Obligations
The DPDP Act does not articulate Principles of Processing, or Fair Information Practice Principles, but the content of several of its provisions put emphasis on purpose limitation (as explained in previous sections of the blog) and on the principle of accountability.
Section 8 of the Act includes multiple obligations for data fiduciaries, all under an umbrella expectation in paragraph 1 that they are “responsible for complying” with the provisions of the Act and any subsequent implementation rules, both regarding processing undertaken by the data fiduciary and by any processor on its behalf. This specification echoes the GDPR accountability principle. In addition, data fiduciaries are under an obligation to implement appropriate technical and organizational measures to ensure the effective implementation of the law.
Data security is of particular importance, considering that data fiduciaries must both take reasonable security safeguards to prevent personal data breaches, and notify the Board and each affected party if such breaches occur. The details related to modalities and timeline of notification will be specified in subsequent implementation rules.
A final obligation of data fiduciaries to highlight is the requirement they establish a “readily available” mechanism for redressing “grievances” by data principals in a timely manner. The “grievance redress” mechanism is of utmost importance, considering that data principals cannot address the Board with a complaint until they “exhaust the opportunity of redressing” the grievance through this mechanism (Section 13(3)). The Act leaves determination of the time period for responding to grievances to delegated legislation, and it is possible that there may be different time periods for different categories of companies.
Fiduciaries Have a Mandate to Verify Parental Consent for Processing Personal Data of Minors under 18
The DPDP Act creates significant obligations concerning the processing of children’s personal data, with “children” defined as minors under 18 years of age, without any distinguishing sub-category for older children or teenagers. As a matter of principle, data fiduciaries are forbidden to engage in any processing of children’s data that is “likely to cause any detrimental effect on the well-being of the child.”
Data fiduciaries are under an obligation to obtain verifiable parental consent before processing the personal data of any child. Similarly, consent must be obtained from a lawful guardian before processing the data of a person with disability. This obligation, which is increasingly common to privacy and data protection laws around the world, may create many challenges in practice. A good resource for untangling its complexity and applicability is FPF’s recently published report and accompanying infographic – “The State of Play: Is Verifiable Parental Consent Fit For Purpose?”
Finally, the Act also includes a prohibition on data fiduciaries engaging in tracking or behavioral monitoring of children, or targeted advertising directed at children. Similar to many other provisions of the Act, the government may issue exemptions from these obligations for specific classes of fiduciaries, or may even lower the age of digital consent for children when their personal data is processed by designated data fiduciaries.
The Act Creates a Data Protection Board to Enforce the Law, But Reserves Regulatory Powers For the Government
The DPDP Act empowers the Government to establish the Board as an independent agency that will be responsible for enforcing the new law. The Board will be led by a Chairperson and will have Members appointed by the Government for a renewable two-year mandate.
The Board is vested with the power to receive and investigate complaints from data principals, but only after the principal has exhausted the internal grievance redress mechanism set up by the relevant data fiduciaries. The Board can issue binding orders against those who breach the law, can direct urgent measures to remediate or mitigate a data breach, imposing financial penalties and direct parties to mediation.
While the Board is granted “the same powers as are vested in a civil court” – including summoning any person, receiving evidence, and inspecting any documents (Section 28(7)), the Act specifically excludes any access to civil courts in the application of its provisions (Section 39), creating a de facto limitation on effective judicial remedy similar to the relief provided in Article 82 GDPR. The Act grants any person affected by a decision of the Board the right to pursue an appeal in front of an Appellate Tribunal, which is designated the Telecom Disputes Settlement and Appellate Tribunal established under other Indian law.
Penalties for breaches of the law have been stipulated in a Schedule attached to DPDP Act and range from the equivalent in rupees of USD $120 to USD $30.2 million. The Board can determine the penalty amount from a preset range based on the offense.
However, the Board does not have the power to pass regulations to further specify details related to the implementation of the Act. The Government is conferred broad discretion in adopting delegated legislation to further specify the provisions of the Act, including clarifying modalities and timelines for fiduciaries to respond to requests from data principals, the requirements of valid notice for obtaining a data principal’s consent for processing of data, details related to data breach notifications, and more. The list of operational details that may be specified by the Government in subsequent rules is open-ended and detailed in Section 40(2)(a) to (z). Subsection (z) of this provision provides a catch-all permitting the Central Government to prescribe rules on “any other matter” related to the implementation of the Act.
In practice, it is expected that it will take time for the new Board to be established and for rules to be issued in key areas for compliance.
Besides rulemaking power, the Central Government has another significant role in the application of the law. Pursuant to Section 36, it can require any information (including presumably personal data) that it wants (or “call for”) from the Board, data fiduciaries, and “intermediaries” as defined by the IT Act. No further specifications are made in relation to such requests, other than that they must be made “for the purposes of the Act.” This provision is broader and subject to fewer restrictions than provisions on data access requests in the existing IT Act and its subsidiary rules.
Additionally, the Central Government may also order or direct any governmental agency and any “intermediary” to block information for access by the public “in the interests of the general public.” To issue such an order, the Board will need to have sanctioned the data fiduciary concerned at least twice in the past, and the Board must advise the Central Government to issue such an order. An order blocking public access may refer to “any computer resource” that enables data fiduciaries to offer goods or services to data principals within the territory of India. While it is now common among modern comprehensive data protection laws around the world for independent supervisory authorities to order erasure of personal data unlawfully processed, or to order international data transfers or sharing of personal data to cease if conditions of the law are not met, these provisions of the DPDP Act are atypical because the orders will come directly from the Government, and also because they more closely resemble online platform regulation than privacy law.
Exceptions for Publicly Available Data And Processing for Research Purposes Are Notable for Training AI
Given that this law comes in the midst of a global conversation about how to regulate artificial intelligence and automated decision-making, it is critical to highlight provisions in the law that seem directed at facilitating development of AI trained on personal data. Specifically, the Act excludes from its application most publicly available personal data, as long as it was made publicly available by the data principal – for example, a blogger or a social media user publishing their personal data directly – or by someone else under a legal obligation to publish the data, such as personal data of company shareholders that regulated companies must publicly disclose by law.
Additionally, the Act exempts the processing of personal data necessary for research or statistical purposes (Section 17(2)(b)). This exemption is extremely broad, with only one limitation in the core text: the Act will still apply to research and statistical processing if the processing activity is used to make “any decision specific to the data principal.”
There is only one other instance in the DPDP Act where processing data to “make decisions” about a data principal is raised. Data fiduciaries are under an obligation to ensure the “completeness, accuracy and consistency” of personal data if it is used to make a decision that affects the data subject. In other words, while the Act does not provide for a GDPR-style right not to be subject to automated decision-making, it does require that when personal data are used for making any individual decisions, presumably including automated or algorithmic decisions, such data must be kept accurate, consistent and complete.
Additionally, the DPDP Act remains applicable to any processing of personal data through AI systems, if the other conditions of the law are met, given the broad definitions of “processing” and of “personal data.” Further rules adopted by the Central Government or other notifications may provide more guidance in this regard.
Notably, the Act does not exempt processing of personal data for journalistic purposes, a fact criticized by the Editors’ Guild of India. In previous versions of the Bill, especially the expert version spearheaded by Justice Srikrishna in 2017, this exemption was present. It is still possible that the Central Government will address this issue through delegated legislation.
Key Takeaways and Further Clarification
India’s data protection Act has been in the works for a significant period of time and the passage of the law is a welcome step forward after the recognition of privacy as a fundamental right in India by the Supreme Court in its landmark Puttaswamy judgment.
While the basic structure of the law is similar to many other global laws like the GDPR and its contemporaries, India’s approach has its differences, such as more limited grounds of processing, wide exemptions for government actors, regulatory powers for the government to further specify the law and to exempt specific fiduciaries or classes of fiduciaries from key obligations, no baked-in definition or heightened protection for special categories of data, and the rather unusual inclusion of powers for the Government to request access to information from fiduciaries, the Board and “intermediaries”, as well as to block access by the public to specific information in “computer resources”.
Finally, we note that many details of the Act are still left to be clarified once the new Data Protection Board of India is set up and further rules for the specification of the law are drafted and officially notified.
Editors: Lee Matheson, Dominic Paulger, Josh Lee Kok Thong
FPF at Singapore PDP Week 2023: Navigating Governance Frameworks for Generative AI Systems in the Asia-Pacific
Authors: Cheng Kit Pang, Elena Guañuna, Alistair Simmons, and Matthew Rostick
Cheng Kit Pang, Elena Guañuna, Alistair Simmons, and Matthew Rostick are FPF Global Privacy Interns.
From July 18 to July 21, 2023, the Personal Data Protection Commission (PDPC) of Singapore held its annual Personal Data Protection Week (PDP Week), which overlapped with the IAPP’s Asia Privacy Forum 2023.
The Future of Privacy Forum (FPF)’s flagship event during PDP Week was a roundtable on the governance implications of generative AI systems in the Asia-Pacific (APAC) region. In organizing this event together with the PDPC, FPF brought together over 80 participants from industry, academia, the legal sector, and international organizations, as well as regulators from across the APAC region, Africa, and the Middle East.
FPF Roundtable on Governance of Generative AI Systems in APAC
On July 21, FPF organized a high-level closed-door roundtable, titled “Navigating Governance Frameworks for Generative AI Systems in the Asia-Pacific.” The roundtable explored the issues raised by applications of existing and emerging AI governance frameworks in APAC to generative AI systems.
Dominic Paulger (Policy Manager, FPF APAC) kicked off the roundtable with a presentation on the existing governance and regulatory frameworks that apply to generative AI systems in the APAC region. The presentation highlighted that to date, most major APAC jurisdictions have opted for “soft law” approaches to AI governance, such as developing ethical frameworks and voluntary governance frameworks, rather than “hard law” approaches, such as enacting binding regulations. However, the presentation also explained that China is an exception to this rule and has been active in enacting regulations targeting specific AI technologies, such as deep synthesis technologies and most recently, generative AI. In addition, even if they do not specifically target Generative AI, the comprehensive data protection laws enacted in most jurisdictions in the region are also applicable to how these types of computer programs are trained and generally process personal data.
The presentation was followed by three hours of discussion, facilitated by Josh Lee Kok Thong (Managing Director, FPF APAC). The discussions were first initiated by firestarters from industry, regulators, and academia:
Denise Wong (Deputy Commissioner, PDPC);
Jeth Lee (Chief Legal Officer, Microsoft Singapore);
Arianne Jimenez (Head of Privacy and Data Policy, Engagement, Meta); ; and
Jason Allen Grant (Director, Centre for AI & Data Governance, and Associate Professor of Law at Singapore Management University Yong Pung How School of Law).
Turning to the wider roundtable discussion, participants highlighted the fast pace of developments in generative AI technology and hence, the importance of adopting an agile and future-proof approach to governance. Participants also identified that compared with other forms of AI technology, generative AI systems were more likely to raise challenges in addressing unseen bias in very large, unstructured data sets and “hallucinations” (generated output that is grammatically accurate but nonsensical or factually inaccurate).
To address these issues, participants highlighted the importance of developing standards and metrics for evaluating the safety of generative AI systems and for measuring the effectiveness of achieving desired outcomes. Participants also called for efforts to educate users on generative AI systems, including the capabilities, limits, and risks of these technologies.
Regarding regulation of generative AI, participants were generally in favor of an incremental approach to the development of governance principles for generative AI systems in the region – allowing actors in the AI value chain to explore ways to operationalize existing AI principles and apply existing governance frameworks to the technology – rather than enacting “hard law” regulations.
Participants also agreed on the need for AI governance principles to account for the three basic layers of the AI technology stack as different policy considerations apply at each of these levels, namely:
The infrastructure layer, which includes the computing hardware, such as central processing units (CPUs) and graphics processing units (GPUs), that is used to train models and the data centers where this hardware is housed;
The model layer, which includes the training and operation of generative AI models like Bard, GPT, LLaMa, Stable Diffusion, and so on; and
The application layer, which includes software applications built on top of generative AI models and the different use cases in which these applications are deployed.
Several participants also raised that at the ecosystem level, it would be important for stakeholders to develop a common or standardized set of terminologies or taxonomies for key concepts in generative AI technology, such as “foundation models” or “large language models” (LLMs).
Some participants also called for greater collaboration between stakeholders, and a multidisciplinary approach to governance of generative AI systems and global alignment when developing best practices.
Photos: Participants from FPF Roundtable on Navigating Governance Frameworks for Generative AI Systems in the Asia-Pacific, 7/21/2023. Photos courtesy of the PDPC.
Other FPF Activities during PDP Week
IAPP Asia Privacy Forum 2023
On July 20, FPF organized an IAPP panel discussion titled “Unlocking Legal Bases for Processing Personal Data in APAC: A Practical Guide,” which built on FPF’s year-long research project on consent and other legal bases for processing personal data in the APAC region – the final report of which was released in November 2022.
Moderator Josh Lee Kok Thong led the discussion, in which panelists Deputy Commissioner Denise Wong, Deputy Commissioner Leandro Y. Aguirre, Arianne Jimenez, and David N. Alfred (Co-Head, Data Protection, Privacy & Cybersecurity Practice, Drew & Napier) explained the challenges faced by practitioners and regulators in addressing differing data requirements for consent and alternatives like “legitimate interests” in APAC data protection laws.
Photo: FPF Panel on Unlocking Legal Bases for Processing Personal Data in APAC, July 20, 2023.
FPF’s APAC office was also represented at two further panels during IAPP Asia Privacy Forum 2023:
Josh moderated a panel titled “Privacy-First Future: Partnering with Industry and Regulators for an Open Internet” focusing on the PDPC’s newly launched PETs x Privacy Sandbox initiative to reduce cross-site and cross-app tracking.
Josh alsospoke on another panel titled “From DPO to Data Ethics Officer, Don’t Fall Behind!” which explored whether data protection officers (DPOs) are suited to fulfill the emerging role of a “data ethics officer” and how DPOs can raise data ethics issues to their company boards.
FPF Training on EU AI Act
On the sidelines of PDP Week, FPF held its inaugural in-person FPF Training session in the APAC region. The closed-door training session, which focused on the forthcoming EU AI Act and its impact on the APAC region, was held on July 20 and was conducted by Katerina Demetzou (Senior Counsel for Global Privacy, FPF) with interventions from Vincenzo Tiani from his experience of advising Members of the European Parliament (MEPs) on drafting the EU AI Act. The training provided a detailed analysis of the draft AI Act and explained the lifecycle of AI systems and the law-making process in the EU. The training drew close to 20 attendees comprising regulators and representatives from industry and the legal sector.
Photo: FPF Training on the EU AI Act, 7/20/2023
Conclusion
This was the second time that FPF organized events around PDP Week since the launch of FPF’s APAC office in 2021. The week’s events enabled FPF APAC to foster collaborative dialogues among regulators, industry, academia, and civil society from the APAC region and draw links with the EU, and the US. FPF is grateful for the support of the PDPC and IAPP in organizing these activities.
Edited by Dominic Paulger and Josh Lee Kok Thong
FPF Files Comments for the FTC Health Breach Notification Rule Addressing Specific Definitions and Clarity of Scope
On August 8th, the Future of Privacy Forum (FPF) filed comments with the U.S. Federal Trade Commission (the Commission) regarding the Notice of Proposed Rulemaking (NPRM) to clarify the scope and application of the Health Breach Notification Rule (HBNR).
The HBNR was promulgated in 2009 as part of the American Recovery and Reinvestment Act as a breach of security rule. Recent complaints brought by the Commission, GoodRx and Easy Healthcare, were the inaugural and second application of the HBNR and indicated a novel range of alleged privacy breaches rather than traditional security breaches. The cases indicated a shift in the interpretation of “breach of security” by the Commission that drew many proto-typical practices into scope. The NPRM seeks to clarify this broadened scope which has amalgamated traditional breaches of security with nascent breaches of privacy. To draw out and address key issues in the NPRM and the Commission’s considerations, we recommended that the Commission consider the nuance of definitions and address the complexities of breach by specifically:
Define a Standard for Identifiability for “PHR identifiable health data” to Clearly Expand Protections for a Broad Spectrum of Personal Information
Define “Relates to” to Include the Creation of Health-Related Inferences from a Wide Range of Routine Commercial Datasets, While Establishing Clear Obligations for Businesses
Establish Clear Guidelines for Intentional Data Sharing that Does Not Require Affirmative Consent
Ensure that the Rule Contains “Good Faith” Exceptions for Merely Technical Violations
Further Define “Breach of Security” to Clarify Where the Commission May Take Enforcement Action
FPF’s full comments to the Commission are available here.
FPF Releases Generative AI Internal Policy Checklist To Guide Development of Policies to Promote Responsible Employee Use of Generative AI Tools
Today, the Future of Privacy Forum (FPF) releases the Generative AI for Organizational Use: Internal Policy Checklist. With the proliferation of employee use of generative AI tools, this checklist provides organizations with a powerful tool to help revise their internal policies and procedures to ensure that employees are using generative AI in a way that mitigates data, security, and privacy risks, respects intellectual property rights, and preserves consumer trust.
The Checklist draws from a series of consultations with practitioners and experts from over 30 cross-sector companies and organizations to understand current and anticipatory employee use of generative AI tools, benefits and harms, AI governance, and measures taken to protect company data and infrastructure. The conversations focused on any generative AI guidelines, policies, and procedures that had been implemented to govern employees’ use of generative AI tools.
From those discussions, we learned that organizations have broadly varied use cases for generative AI and, therefore, significant variation in generative AI policies. Some organizations have enacted outright bans for generative AI tools without prior approval, while others have created restrictions for the use of generative AI, and still, others have yet to develop express policies and procedures on employee use of generative AI. The Internal Policy Checklist for Generative AI is intended to serve as a guidance document no matter what stage of the process an organization is in. It may be used as a starting point to help kick off the development of internal generative AI policies or as a final check to ensure an organization has provided comprehensive and robust guidelines for their teams.
“It is imperative that both organizations and their employees understand the benefits and risks of generative AI tools, and that organizations have appropriate safeguards in place to support responsive and ethical use,” said Amber Ezzell, AI policy counsel at FPF and author of the checklist. “Employee use of generative AI tools is inevitable and may bring new and unexpected benefits to employers as employees find ways to be more productive and creative in even the most mundane tasks. Developing thoughtful generative AI policies is essential to ensure you’re well prepared for the changing way of work.”
Use in Compliance with Existing Laws and Policies for Data Protection & Security. Designated teams or individuals should revisit internal policies and procedures to ensure that they account for planned or permitted uses of generative AI. Employees must understand that relevant current or pending legal obligations apply to the use of new tools.
Employee Training and Education. Identified personnel should inform employees of the implications and consequences of using generative AI tools in the workplace, including providing training and resources on responsible use, risk, ethics, and bias. Designated leads should provide employees with regular reminders of legal, regulatory, and ethical obligations.
Employee Use Disclosure. Organizations should provide employees with clear guidance on when and whether to use organizational accounts for generative AI tools, as well as policies regarding permitted and prohibited uses of those tools in the workplace. Designated leads should communicate norms around documenting use and disclosing when generative AI tools are used.
Outputs of Generative AI. Systems should be implemented to remind employees to verify outputs of generative AI, including for issues regarding accuracy, timeliness, bias, or possible infringement of intellectual property rights. Organizations should determine whether and to what extent compensation should be provided to those whose intellectual property is implicated by generative AI outputs. When generative AI is used for coding, appropriate personnel should check and validate outputs for security vulnerabilities.
Old Laws & New Tech: As Courts Wrestle with Tough Questions under US Biometric Laws, Immersive Tech Raises New Challenges
Extended reality (XR) technologies often rely on users’ body-based data, particularly information about their eyes, hands, and body position, to create realistic, interactive experiences. However, data derived from individuals’ bodies can pose serious privacy and data protection risks for people. It can also create substantial liability risks for organizations, given the growing volume of lawsuits under the Illinois Biometric Information Privacy Act (BIPA) and scrutiny of biometric data practices by the Federal Trade Commission (“FTC” or “Commission”) in their recent Policy Statement. At the same time, there is considerable debate and lack of consensus about what counts as biometric data under existing state privacy laws, creating significant uncertainty for regulators, individuals, and organizations developing XR services.
This blog post explores the intersection of US biometric data privacy laws and XR technologies, particularly whether and to what extent specific body-based data XR devices collect and use may be considered “biometric” under various data protection regimes. We observe that:
Face templates and iris scans used to authenticate an individual’s identity are regulated biometrics, therefore those use cases in XR are covered by biometric laws.
Laws with broad definitions of biometrics may apply to systems that use face detection, as seen in emerging case law from Illinois regarding virtual try-on XR applications.
Organizations have taken steps that reduce their liability risk regarding face-based biometric systems, including by minimizing collection of identifying data or processing biometric data on individuals’ devices.
Other body-based data not used for identification in XR, like eye-tracking and voice analysis, may also be considered “biometric” if the technology and data are capable of identifying an individual.
A. Face Templates, Hand Scans, and Iris Scans Used to Authenticate an Individual’s Identity Are Regulated Biometrics, Therefore User Authentication in XR is Covered by Biometric and Comprehensive Privacy Laws
With the exception of CUBI (and to a certain extent, BIPA), most biometric and comprehensive data privacy statutes tie their definitions of “biometric data” to identification, meaning the laws are intended to regulate unique physiological characteristics that entities use to identify an individual. Generally, each biometric and comprehensive law focuses on five forms of biometric data: retina or iris scan, fingerprint, voiceprint, hand scan, and face scan. BIPA, in particular, applies to “biometric identifiers,” defined as a “retina or iris scan, fingerprint, voiceprint, or scan of hand or face geometry,” as well as “biometric information,” which includes “any information…based on an individual’s biometric identifier used to identify an individual.” Therefore, any entity that uses technology to scan an individual’s iris, finger, hand, or face to uniquely identify an individual (1:many) or authenticate their identity (1:1) must comply with BIPA’s requirements, unless they fall within one of BIPA’s exemptions or exclusions. The same conclusion applies to CUBI, BPPA, and comprehensive data privacy laws.
XR devices often use iris, face, or hand scans to authenticate a user’s identity to log in to their profile or enable in-app payments. Much like computers or smartphones, more than one user may use a single XR device, so authenticating the specific person using the device at a given time allows for more personalization and secure transactions. As a result, iris or face authentication systems in XR devices are likely covered by U.S. biometric and comprehensive data privacy laws. The laws typically require organizations to obtain user consent before enrolling individuals in this sort of biometric XR system, and BIPA has potentially thorny provisions requiring “written consent,” which can be challenging to implement for many XR applications. The face and eye scans that XR technologies use for authentication may also be considered “sensitive data” under comprehensive data privacy laws, such as the California Privacy Rights Act (CPRA) or the Connecticut Data Privacy Act, requiring organizations to provide individuals opt-out rights, including the right to opt out of data sales and other transfers.
Most XR authentication technologies employ live capture of biometrics. Iris, face, or hand scans are captured in real-time when an individual first enrolls, and subsequent scans are likewise captured in real-time when the individual authenticates their identity to the device or app. These scenarios are typically covered by biometrics laws as described above. However, there is some uncertainty regarding biometric laws’ application to XR devices that create a biometric template from non-real-time photos, videos, and audio recordings. Most biometric and comprehensive privacy laws exclude photos, videos, and audio recordings from the scope of “biometric data” to varying degrees (with the exception of CUBI and the CPRA). Utah and Virginia’s comprehensive privacy laws, for example, broadly exempt photographs “or any data generated therefrom” from coverage, making their biometric regulations perhaps less likely to apply to photographic scanning. But case law under BIPA shows that these provisions may not exclude “biometric templates” derived from non-real-time photos, videos, or audio recordings. In Shutterfly v Monroe, the United States District Court for the Northern District of Illinois concluded that narrowly reading “biometric identifier” only to mean real-time scans would swallow the intent of the law, thus photographic scanning to create “templates” were still within scope. Laws like the Connecticut Data Privacy Act (CTDPA) and the final rules for the Colorado Privacy Act (CPA) that do not exclude photos or data generated from these sources if an entity uses them for identification purposes, or CUBI, which contains no exemptions for photographs at all, are likely to follow this analysis. The FTC’s conception of biometric information similarly and explicitly encompasses photos, videos, audio recordings, and certain data derived from these sources, making it likely that most regulators will still consider “biometric templates” created from photographic scanning subject to applicable biometric regulations.
B. Laws with Broad Definitions of Biometrics May Apply to Systems that Use Face Detection, as Seen in Emerging Case Law from Illinois Regarding Virtual Try-On XR Applications
Despite most laws’ goal to regulate biometric data that is uniquely identifying, several statutes’ text can be interpreted to apply to biometric technologies that merely distinguish a face from other objects or analyze facial characteristics, without identifying a particular individual. Depending on a privacy law’s definition of “biometric data,” courts may hold that the term regulates technologies that utilize data derived from an individual’s face, eyes, or voice even when they are not used for identification purposes. In XR, devices may use inward-facing cameras to conduct facial analysis for non-identification purposes, such as rendering expressive avatars. Augmented reality (AR) products like “virtual try-on” may also use facial analysis for people to visualize how different products – like eyeglasses – might look on them. Like many other XR applications, VTO primarily uses facial scans to detect and correctly align the product with an individual’s physical features, rather than for identification purposes.
Some laws with broad definitions can apply to these non-identification technologies unless a specific exception applies. CUBI does not require “biometric identifiers” to uniquely identify an individual, which has prompted the Texas Attorney General to claim that CUBI applies to the capture of face geometry regardless of whether an entity uses these facial maps for individual identification. The FTC’s conception of biometric technologies also broadly encompasses “all technologies that use or purport to use biometric information for any purpose.” But most notably, BIPA is complex because its definition of “biometric identifiers” does not explicitly require that the data be used for identification (in contrast to the statute’s definition of “biometric information,” which does require identification). As a result, Illinois courts have largely found that any facial scan may create a “biometric identifier,” such as with doorbell cameras, photo grouping, and Snapchat filters. This is true even when that technology’s facial scan feature was not used to identify the individual in the photo or video frame.
Recent BIPA lawsuits brought against companies that offer (VTO) illustrate how broad biometric laws might apply to XR devices that use facial analysis. In Theriot v. Louis Vuitton North America, Inc., a federal court permitted BIPA claims to proceed against Louis Vuitton’s VTO sunglasses application, finding that the technology’s use of facial scans was analogous to BIPA case law holding that face scans derived from photographs constitute biometric identifiers. Other VTO cases have had similar outcomes. Only VTO technology used for healthcare-related purposes, such as trying on prescription eyeglasses, have been found by courts to be outside the scope of BIPA. But this result did not rest on BIPA’s overall definition of biometric data, but rather arose from a narrow exception for “information captured from a patient in a health care setting.” So BIPA may not apply to medical providers’ use of XR apps or other immersive technologies, such as brain computer interfaces (BCIs), for diagnostic purposes, but BIPA’s coverage of non-identifying, non-medical uses remains a source of substantial confusion. This confusion undermines individuals’ understanding of their privacy rights and presents liability risks for organizations.
C. Organizations May Reduce their Liability Risk by Minimizing Collection of Identifying Data or Processing Biometric Data on Individuals’ Devices
Some organizations have taken steps to limit their liability risks by minimizing the collection of identifying data or processing biometric data on individuals’ devices. Case law suggests that some facial detection technologies fall outside the scope of BIPA and other biometric regulations if (1) there is no mechanism for the technology to retain facial scans or link scans to a user’s individual identity or account; and/or (2) all of the data is stored on-device.
First, in Daichendt and Odell v. CVS Pharmacy, the Northern District of Illinois dismissed a case against CVS for its passport photosystem, which scans facial geometry in photos to confirm that they meet government requirements for passports (e.g., a person’s eyes are open, their mouth is closed and not smiling, and eyeglasses are not present). The court held that the plaintiffs failed to allege that CVS’ photosystem enabled CVS to determine their identities, nor did the plaintiffs provide CVS “with any information, such as their names or physical or email addresses, that could connect the voluntary scans of face geometry with their identities.”
Separately, in Apple v. Barnett, the Illinois’ appellate court held that Apple was not subject to BIPA requirements regarding their Face ID on iPhone because the company was not “collecting” or “possessing” users’ biometric data since the data was completely stored on the device and never stored on Apple servers. Thus, XR devices that do not retain facial scans that can link to users’ accounts, or only store data on-device (such as Apple’s recently announced Vision Pro) may be out of scope of even some of the broadest biometrics laws.
D. Eye-tracking and Voice Analysis May Also be Considered “Biometric” if the Technology and Data are Capable of Identifying an Individual
In addition to face-based biometric technologies, most XR devices also use other forms of body-based detection or characterization systems for device functionality, such as voice analysis and eye-tracking. As seen with facial detection, these features are developed to detect or create predictions regarding bodily characteristics or behavior, but the subject is typically not identifiable and PII is typically not retained. For example, XR devices often contain microphones to capture a user’s voice and surroundings, which can enable voice commands, verbal interactions with other users, spatial mapping, and realistic sound effects. XR devices may also maintain inward-facing cameras that collect data about a user’s gaze—where they look and for how long—to enable eye tracking. This may be used to improve graphics and allow for more expressive avatars, including avatars that can display microexpressions.
Whether these systems that collect voice or gaze data are covered by biometric or comprehensive data privacy laws may depend on whether an organization can use the technology to identify an individual, even if not used in that capacity. As seen in CVS Pharmacy, many Illinois courts focus on the capacity of the technology to identify an individual. As an initial matter, biometric and comprehensive privacy laws typically apply to “voiceprints,” and not voice recordings. As stated by the Illinois Attorney General, “a voiceprint, which is a record of mechanical measurement, is not the same as a simple recording of a voice.”
However, the line between a voice recording and a voiceprint is blurry, particularly as it relates to the gray area of natural language processing (NLP)—a kind of artificial intelligence (AI) that can use audio to understand, interpret, and manipulate language. In Carpenter v. McDonald’s Corp., the U.S. District Court for the Northern District of Illinois found that McDonald’s drive-through voice assistant technology could be used for identification purposes, and thus could be considered a “voiceprint” under BIPA, since the technology’s patent application states that the technology may capture voice characteristics “like accent, speech pattern, gender, or age for the purpose of training the AI.” In a similar ongoing case against Petco, an Illinois federal judge permitted BIPA claims to proceed regarding employee voice data, stating “[w]hat matters [at the dismissal stage] is not how defendant’s software actually used plaintiffs’ data, but whether the data that Petco gathered was capable of identifying [the employees].” As a result, if an XR device captures vocal characteristics that are capable of unique identification, certain voice data may be considered a “voiceprint” under BIPA. This analytical framework will likely apply to jurisdictions that define biometric data to include biological characteristics that have the potential to identify an individual, such as in the final rules under the Colorado Privacy Act regarding biometric identifiers, or under the FTC’s Policy Statement on Biometric Information.
Whether privacy laws apply to gaze data, however, is even less clear. BIPA lawsuits against online exam proctoring services, autonomous vehicles, and “smart advertising screens” suggest that eye-tracking could be a biometric identifier under BIPA, even if not used for identification. In each of these cases, the technology conducted eye-tracking to determine where a user was looking—whether on the screen, the road, or in the store—but did not identify the individual. Instead, these technologies made inferences about whether someone may be cheating, not paying attention to the road, or what product they were looking at. Plaintiffs in these cases argue that eye-tracking is part of the technology’s collection and analysis of facial geometry, thus making it a “biometric identifier” under BIPA.
Unfortunately, state and federal courts in Illinois have not analyzed whether and to what extent eye tracking, without additional face analysis, constitutes a biometric identifier, nor whether it is a subset of facial analysis. Rather, most cases proceed based solely on the software’s overall facial analysis features, if at all. If courts are prone to equate facial detection scans to “facial geometry,” and voice analysis to “voiceprints,” they may also conflate eye tracking with “a retina or iris scan,” and thus treat eye tracking as a biometric identifier. Or they may follow the BIPA plaintiffs’ analysis, lumping eye-tracking into facial analysis as “facial geometry.” Alternatively, courts could characterize eye tracking as altogether separate from BIPA’s “facial geometry” and “retina or iris scan” categories. In any event, like with voice analysis, if an XR device collects gaze data that could be used for identification purposes, laws with broad biometrics definitions will apply, while other laws that have narrower definitions focused on the data or technology’s current use, may exclude the technology.
Takeaways
Statutory language and court opinions vary in how they define and/or apply to biometric data and identifiers. Though the plain text of most U.S. biometric and comprehensive data privacy laws tie their definition of a “biometric” to the identification of an individual, some laws may be more broadly applied to technologies that use body-based data for non-identification purposes. While most of the body-based data XR collects is not used for identification, litigation brought under BIPA and other state laws suggest that lawmakers and judges may consider certain kinds and uses of such data—for example, AR “facial scans,” eye tracking, and voice— to be biometrics. Whether this will be the case (or continue to be the case) depends on how policymakers draft these laws, and how courts, enforcement bodies, and other parties to litigation interpret statutes regulating biometrics.
Insights into Brazil’s AI Bill and its Interaction with Data Protection Law: Key Takeaways from the ANPD’s Webinar
Authors: Júlia Mendonça and Mariana Rielli
The following is a guest post to the FPF blog by Júlia Mendonça, Researcher at Data Privacy Brasil, and Mariana Rielli, Institutional Development Coordinator at Data Privacy Brasil. The guest blog reflects the opinion of the authors only. Guest blog posts do not necessarily reflect the views of FPF.
On July 6, 2023, the Brazilian National Data Protection Authority (ANPD) held a webinar event entitled: The interplay between AI regulation and data protection. The dialogue unfolded in the broader context of developments in AI regulation in Brazil which has, as its main drivers, the bills that propose a Regulatory Framework for Artificial Intelligence in the country. The bills were jointly analyzed by a Commission of 18 jurists appointed by the Federal Senate, which promoted meetings, seminars, and public hearings to substitute them with a new draft proposal. At the beginning of May, the draft produced by the Commission was transformed into a new bill that is currently going through the legislative process: Bill PL nº2338 (AI draft bill).
The ANPD, noting the need to harmonize any upcoming AI regulation with the existing data protection regime (as well as future enforcement matters), organized this webinar, in addition to having published a preliminary analysis of the AI draft bill. The discussions during the webinar offer a glimpse into the AI lawmaking and policymaking in Brazil, one of the largest jurisdictions in the world – one that is also covered by a general data protection law applicable to personal data processed in the context of an AI system. This brief blog post outlines the main topics discussed during the event, particularly in relation to the interplay between the current AI draft bill and Brazil’s General Data Protection Law (LGPD).
The webinar’s opening welcomed Waldemar Gonçalves (President, ANPD, Brazil), Eduardo Gomes (Senator of the Republic, Brazil), and Estela Aranha, (Special Advisor, Ministry of Justice and Public Security, Brazil). The panel that followed was formed by representatives of the National Data Protection Council (CNPD) – a multisectoral advisory body, part of the ANPD structure – namely, Ana Paula Bialer (Founding Partner, Bialer Falsetti Associados, Brazil), Bruno Bioni, (Director and Founder, Data Privacy Brasil), Fabrício da Mota (Vice President, Conselho Federal da OAB, Brazil), and Laura Schertel (Visiting researcher, Goethe Universität Frankfurt; and private law Professor and lawyer, Brazil/EU).
Key representatives highlight the need for ongoing harmonization between AI regulation and data protection law in Brazil
As the President of the ANPD, Waldemar Gonçalves highlighted the Authority’s ongoing work on the AI agenda, noting that data protection rules under the LGPD are closely interconnected with those provided for in the AI draft bill, such as with regard to the right to information. With such similarities in mind, Gonçalves noted the need for harmonization between different tools, such as the Data Protection Impact Assessment (DPIA) and the Algorithm Impact Assessment (AIA).
Another initiative of the ANPD highlighted by Gonçalves as relevant to the AI agenda and the current AI regulatory efforts was the technical agreement between the Authority and the Latin American Development Bank (CAF), which will include a regulatory sandbox pilot program on data protection and AI.
ANPD’s current president closed his remarks recalling the various recent cases in which data protection authorities around the world have spoken out on issues concerning AI-based systems, thereby reinforcing the importance of the ANPD in assuming an active role in this discussion. Eduardo Gomes, rapporteur of the AI draft bill, started from the same premises to support the efforts with the president of the Senate, Rodrigo Pacheco. In addition to reinforcing the importance of work of the Commission of Jurists in laying the groundwork for the debate in Brazil, he also recognized the need to foster other opportunities to “mature the subject.”
Concluding the opening panel, Estela Aranha focused her presentation on the topic of algorithmic discrimination in the context of the interplay between AI and existing data protection norms. Aranha mentioned examples with regards to data mining and how the resulting massive collection of data can generate the most varied risks, including risks of discrimination, and can go beyond the most obvious examples of sensitive and inferred data. The relevance of this specific point in the debate stems from the fact that the proposed AI draft bill is quite detailed, both in terms of definitions and obligations created, with regards to direct and indirect discrimination potentially created or enhanced by AI systems in the Brazilian context. Finally, Aranha also reaffirmed the Ministry of Justice’s support for the Bill.
A deeper dive into the proposed AI draft bill and possible future(s) of AI regulation
The following panel focused on a deeper look at the proposed AI draft bill and some of the specific provisions therein. The first panelist, Ana Paula Bialer, highlighted that there is already a robust framework for data protection that grants the data subject greater control over their data, based on the principle of “informational self-determination.” However, Bialer made a point that there may be a certain difficulty in applying the rationale of data protection to AI. Not in the sense that the data used is presumably not protected, but rather that there should be a thorough exercise of extension and “revalorization” of the principles of the LGPD, combined with a review of the set of rights put in place in the context of AI systems.
Already assessing the current draft bill, Bialer also considered that the meaning of a human-centered approach can be different when thinking about different applications of AI in varying socio-economic contexts, exemplifying her reflection through the topic of recruitment and new hires’ selection and the right to full employment in Brazil. Bialer concluded by reaffirming the benefits that can be brought by AI for social and economic development in the country, as well as for the exercise of fundamental rights. In this context, Bialer welcomed the ANPD’s regulatory sandbox initiative and positioned herself more favorably to a strongly risk-based approach to AI regulation.
Bruno Bioni began by emphasizing the importance of having a dose of skepticism with regards to the broader debate – both on AI, and in respect to AI regulation – especially in a scenario where the almost “apocalyptic” narrative around AI continues gaining notoriety. This is important because, in Bioni’s opinion, such discourse may end up underestimating the regulatory tools that already exist. The very field of personal data protection has already provided positive and negative lessons when it comes to an object of regulation that is very plastic and polyvalent, “with a regulatory mission that is transversal and not sectoral.”
Bioni continued by pointing out that the intersection of data protection, AI regulation, and governance is very much related to the idea of a “toolbox” that opens opportunities for a more collaborative, collective regulatory production, relying on companies themselves to participate and to some extent, be rewarded, for example, if they demonstrate a good level of accountability.
Among the various existing tools and how they can support each other, Bioni highlighted Algorithmic Impact Assessments (AIA) and Data Protection Impact Assessments (DPIAs) as documentation that can foster and unfold into the other in such a way as to optimize both. The ANPD has already positioned the DPIA prescribed by the LGPD as an instrument to be better regulated and better standardized, which, for the expert, will be a significant advancement, even in a hypothetical scenario where it takes a long time for an AI regulation to be passed.
According to Bioni, it is for this reason that data protection authorities around the world have led enforcement actions, in the absence of AI laws or authorities created with this specific mission. Bioni concluded his remarks by pointing out that it is essential to think about a more collective or networked governance approach.
Fabrício da Mota Alves focused on the issue of institutional arrangements and of thinking about future legislation inserted in a regulatory environment that is founded on the administrative action of the Brazilian State. Fabrício pondered on the possibility that, following other countries in the world, the ANPD promotes some degree of administrative action (supervisory and sanctioning, in addition to regulation and awareness) related to AI, reinforcing that there is a concern to understand and call for the ANPD to build a very robust regulatory environment. Above all, there is a call for formal protocols so that companies and experts can understand the limits and the scope of ANPD’s actions in this dynamic scenario.
Celebrating the space provided by the webinar as one of the first and most qualified to take place outside of the legislative environment, Alves emphasized that it is imperative that, also in the context of regulating and enforcing AI-related cases (regardless of specific frameworks), the Brazilian ANPD maintains the stance it has adopted so far, with broad public participation, hearings, public consultations, and processes that are open to criticism from all affected sectors.
What’s next for the Brazilian AI bill?
Brazil’s AI draft bill is in its early stages, although it has already been the result of lengthy discussions by the expert committee assigned to prepare a new draft in 2022. There is an expectation that it will now be analyzed by a special committee of parliamentarians designated specifically to debate the Bill, with the prospect of new rounds of public hearings. After the text is approved by the plenary of the Brazilian Senate, the proposal still goes through the Chamber of Deputies, the reviewing house, until a common text is reached, which will then be sanctioned by the President of the Republic.
The whole webinar, in Portuguese, can be watched here.
