Centre for Internet and Society

Pandemic Technology takes its Toll on Data Privacy

Aman Nair and Pallavi Bedi — 2021-06-26T06:52:52Z

The absence of any legal framework has meant these tools are now being used for purposes beyond managing the pandemic.

The article by Aman Nair and Pallavi Bedi was published in the Deccan Herald on June 13, 2021.

People show Arogya Setu App installed in their phones while travelling by special New Delhi-Bilaspur train from New Delhi Railway Station. Credit: PTI File Photo

Jabalpur: A beneficiary shows his certificate on his mobile phone after receiving COVID-19 vaccine dose, at Gyan Ganga College in Jabalpur, Saturday, May 15, 2021. (PTI Photo)

At a time when technology is spawning smart solutions to combat Covid-19 worldwide, India’s digital response to the pandemic has stoked concerns that surveillance could pose threats to the privacy of the personal data collected. Be it apps or drones, there is widespread criticism that digital tools are being misused to share information without knowledge or consent. At the other end of the spectrum, the great urban-rural digital divide is hampering the already sluggish vaccination drive, exposing vulnerable populations to a fast-mutating virus.

Last year, the Centre, states and municipal corporations launched more than 70 apps relating to Covid-19, demonstrating the country’s digital-driven approach to handling the pandemic. Chief among these was the central government’s contact tracing app Aarogya Setu. Launched under the Digital India programme, the app quickly came under scrutiny over data privacy.

As per its privacy policy, Aarogya Setu collects personal details such as name, age, sex, profession and location. As there is no underlying legislation forming its basis, and in the absence of a personal data protection bill, serious privacy concerns regarding the collection, storage and use of personal data have been raised.

The government has attempted to mitigate these concerns with reassurances that the data will be used solely in tracing the spread of the virus. However, recent reports from the Kulgam district of Jammu and Kashmir point to the sharing of application data with police. This demonstrates how easy it is to use personal data for purposes other than which it was collected, and presents a serious threat to citizen privacy.

Though Aarogya Setu was initially launched as ‘consensual’ and ‘voluntary’, it soon became mandatory for individuals to download the app for various purposes such as air and rail travel (this order was subsequently withdrawn) and for government officials. Initially it was also mandatory for the private sector, but this was later watered down to state that employers should, on a ‘best effort basis', ensure that the app is downloaded by all employees having compatible phones. However, the ‘best effort basis’ soon translated into mandatory imposition for certain individuals, especially those working in the ‘gig economy’.

Several states had also launched apps for various purposes ranging from contact tracing of suspected Covid patients to monitoring the movement of quarantined patients. As a report by the Centre for Internet and Society observed, given the attention on Aarogya Setu, most of the apps launched by the state governments escaped scrutiny and public attention.Most of these apps either did not have a privacy policy or the policy was vague and often did not provide important details such as who was collecting the data, the time period for retaining the data and whether personal data could be shared with other departments, most notably, law enforcement.Apart from contact tracing apps, the pandemic also ushered in a wave of other apps and digital tools by the government. These include systems such as drones to check whether people are following Covid-19 norms and facial recognition cameras to report to the police whether someone has broken quarantine. Similar to Aarogya Setu, these tools have also largely been brought about in the absence of a legal and regulatory framework.
The absence of any legal framework has meant these tools are now being used for purposes beyond managing the pandemic.

The government is now planning to use facial recognition technology along with Aadhaar toauthenticate people before giving them vaccine shots.

Aarogya Setu is now linked with the vaccination process. Beneficiaries have been provided an option to register through Aarogya Setu. The pandemic has also provided a means for the government to bring in changes to health policies and introduce the National Health Data Management Policy for the creation of a Unique Health Identity Number for citizens.

Vaccination and digital platforms

The use of digital technology has extended to the vaccination process through the deployment of the Covid Vaccine Intelligence Network (Co-WIN) platform.During the first phase of inoculation, beneficiaries were required to register on the Co-WIN app while in the subsequent phases, registration was to be done on the Co-WIN website. The beneficiary is required to upload a photo identity proof.

While Aadhaar has been identified as one of the seven documents that can be uploaded for this, the Health Ministry has clarified that Aadhaar is not mandatory for registration either through Co-WIN or through Aarogya Setu. However, as per media reports, certain vaccination centres still seem to insist on Aadhaar identity even though beneficiaries may have used another identity proof to register on the Co-WIN website.

It is also pertinent to note that the website did not have a privacy policy till the Delhi High Court issued directions on June 2, 2021. The privacy policy hyperlinked on the Co-WIN app directed the user to the Health Data Policy of the National Health Data Management Policy, 2020.

The vaccination drive has been used as a means to push the health identity project forward as beneficiaries who have opted to provide Aadhaar identity proof have also been provided with a health identity number on their vaccination certificate. It is interesting to note that Co-WIN’s privacy policy now states that if the beneficiary uses Aadhaar as identity proof, it can 'opt' to get a Unique Health Id.However, as a recent report revealed, health identity numbers have already been generated for certain beneficiaries without obtaining consent from them for the purpose.

Have the apps been successful?

One could argue that privacy concerns are a worthwhile tradeoffin order to contain the spread of thepandemic. But it is worth examining how successful these technologies have been. In reality, the use of digital technology at every stage of combating the pandemic has clearly highlighted the extent of our digital divide. As per data from TRAI, there are around 750 million Internet subscribers in India,which is only a little more than half of India’s estimated 1.3 billion citizens — with this gap having a significant impact on the efficacy of the government’s strategies. Aarogya Setu has fallen far short of its goal, of having near universal adoption. It has limited adoption in much of the country. This has severely limited its efficacy in tracing the spread of the virus. Research from Maulana Azad Medical College has cited socio-economic inequalities,educational barriers and the lack of smartphone penetration as being the key causes behind the app’s limited success, pointing back to the digital divide. Moreover, the app has also brought with it a host of associated problems including lateral surveillance and function creep caused by the addition of new features. All of which, along with the previously mentioned privacy concerns, have served to hamper public trust and adoption.

A similar situation is seen in the case of vaccination and the Centre’s Co-WIN web portal. The need for registration, first on the Co-WIN app and later on the Co-WIN web portal, has disproportionately affected those who either have no or limited digital access. Many of them belong to vulnerable groups such as migrant and informal sector workers (mainly from disadvantaged castes), LGBTQIA + individuals, sex workers and both urban and rural poor. These issues have also been acknowledged by the Supreme Court, which raised serious concerns about the government being able to achieve its stated object of universal vaccination.

As the inoculation exercise opened up for the 18-45 age group, it increasingly favoured the urban population who possessed the technological and digital literacy to either create or access a host of tools. One need to only look at the wave of automated CO-WIN bots that arose as soon as the vaccination process was expanded to see how these dynamics manifested.

Ultimately, the digital-driven approach that the governments have adopted has resulted in a number of issues — most notably, data privacy and exclusion. Going forward, government strategies must actively account for these factors and ensure that citize rights are adequately protected.

For more details visit https://cis-india.org/internet-governance/blog/deccan-herald-aman-nair-and-pallavi-bedi-june-13-2021-pandemic-technology-takes-its-toll-on-data-privacy

A Guide to Drafting Privacy Policy under the Personal Data Protection Bill, 2019

shwetar — 2021-09-20T10:34:40Z

The Personal Data Protection Bill, 2019, (PDP Bill) which is currently being deliberated by the Joint Parliamentary Committee, is likely to be tabled in the Parliament during the winter session of 2021.

The Bill in its current form, doesn’t have explicit transitory provisions i.e. a defined timeline for the enforcement of the provisions of the Bill post its notification as an enforceable legislation. Since the necessary subject matter expertise may be limited on short notice and out of budget for certain companies, we intend to release a series of guidance documents that will attempt to simplify the operational requirements of the legislation.

Certain news reports had earlier suggested that the Joint Parliamentary Committee reviewing the Bill has proposed 89 new amendments and a new clause. The nature and content of these amendments so far remain unclear. However, we intend to start the series by addressing some frequently asked questions around meeting the requirements of publishing a privacy notice and shall make the relevant changes post notification of the new Bill. The solutions provided in this guidance document are mostly based on international best practices and any changes in the solutions based on Indian guidelines and the revised PDP Bill will be redlined in the future.

The frequently asked questions and other specific examples on complying with the requirements of publishing a privacy policy have been compiled based on informal discussions with stakeholders, unsolicited queries from smaller organizations and publicly available details from conferences on the impact of the Bill. We intend to conduct extensive empirical analysis of additional queries or difficulties faced by smaller organizations towards achieving compliance post the notification of the new Bill. Regardless, any smaller organizations(NGOs, start-ups etc.) interested in discussing compliance related queries can get in touch with us.

Click to download the full report here. The report was reviewed by Pallavi Bedi and Amber Sinha.

For more details visit https://cis-india.org/internet-governance/blog/shweta-reddy-september-17-2021-a-guide-to-drafting-privacy-policy-under-personal-data-protection-bill

Beyond the PDP Bill: Governance Choices for the DPA

Trishi Jindal and S.Vivek — 2021-11-10T07:32:33Z

This article examines the specific governance choices the Data Protection Authority (DPA) in India must deliberate on vis-à-vis its standard-setting function, which are distinct from those it will encounter as part of its enforcement and supervision functions.

The Personal Data Protection Bill, 2019, was introduced in the Lok Sabha on 11 December 2019. It lays down an overarching framework for personal data protection in India. Once revised and approved by Parliament, it is likely to establish the first comprehensive data protection framework for India. However, the provisions of the Bill are only one component of the forthcoming data protection framework It further proposes setting up the Data Protection Authority (DPA) to oversee the final enforcement, supervision, and standard-setting. The Bill consciously chooses to vest the responsibility of administering the framework with a regulator instead of a government department. As an independent agency, the DPA is expected to be autonomous from the legislature and the Central Government and capable of making expert-driven regulatory decisions in enforcing the framework.

Furthermore, the DPA is not merely an implementing authority; it is also expected to develop privacy regulations for India by setting standards. As such, it will set the day-to-day obligations of regulated entities under its supervision. Thus, the effectiveness with which it carries out its functions will be the primary determinant of the impact of this Bill (or a revised version thereof) and the data protection framework set out under it.

The final version for the PDP Bill may or may not provide the DPA with clear guidance regarding its functions. In this article, we emphasise the need to look beyond the Bill and instead examine the specific governance choices the DPA must deliberate on vis-à-vis its standard-setting function, which are distinct from those it will encounter as part of its enforcement and supervision functions.

A brief timeline of the genesis of a distinct privacy regulator for India

The vision of an independent regulator for data protection in India emerged over the course of several intervening processes that set out to revise India’s data protection laws. In fact, the need for a dedicated data protection regulation for India, with enforceable obligations and rights, was debated years before the Aadhaar, Cambridge Analytica, and Pegasusrevelations captured the public imagination and mainstreamed conversations on privacy.

The Right to Privacy Bill, 2011, which never took off, recognised the right to privacy in line with Article 21 of the Constitution of India, which pertains to the right to life and personal liberty. The Bill laid down express conditions for collecting and processing data and the rights of data subjects. It also proposed setting up a Data Protection Authority (DPA) to supervise and enforce the law and advise the government in policy matters. Upon review by the Cabinet, it was suggested that the Authority be revised to an Advisory Council, given its role under the Bill was limited.

Subsequently, in 2012, the AP Shah Committee Report recommended a principle-based data protection law, focusing on set standards while refraining from providing granular rules, to be enforced through a co-regulatory structure. This structure would consist of central and regional-level privacy commissioners, self-regulatory bodies, and data protection officers appointed by data controllers. There were also a few private members’ bills introduced between 2011 and 2019.

None of these efforts materialised, and the regulatory regime for data protection and privacy remained embedded within the Information Technology Act, 2000, and the Information Technology (Reasonable Security Practices and Procedures and Sensitive Personal Data or Information) Rules, 2011 (SPDI Rules). Though the SPDI Rules require body corporates to secure personal data, their enforcement is limited to cases of negligence in abiding by these limited set of obligations pertaining to sensitive personal information only, and which have caused wrongful loss or gain – a high threshold to prove for aggrieved individuals. Otherwise, the Intermediary Guidelines, 2011 require all intermediaries to generally follow these Rules under Rule 3(8). The enforcement of these obligations is entrusted to adjudicating officers (AO) appointed by the central government, who are typically bureaucrats appointed as AOs in an ex-officio capacity.

By 2017, the Aadhaar litigations had provided additional traction to the calls for a dedicated and enforceable data protection framework in India. In its judgement, the Supreme Court recognised the right to privacy as a fundamental right in India and stressed the need for a dedicated data protection law. Around the same time, the Ministry of Electronics and Information Technology (MeitY) constituted a committee of experts under the chairmanship of Justice BN Srikrishna. The Srikrishna Committee undertook public consultations on a 2017 white paper, which culminated in the nearly comprehensive Personal Data Protection Bill, 2018, and an accompanying report. This 2018 Bill outlined a regulatory framework of personal data processing for India and defined data processing entities as fiduciaries, which owe a duty of care to individuals to whom personal data relates. The Bill provided for the setting up of an independent regulator that would, among other things, specify further standards for data protection and administer and enforce the provisions of the Bill.

MeitY invited public comments on this Bill and tabled a revised version, the Personal Data Protection Bill, 2019 (PDP Bill), in the Lok Sabha in December 2019. Following public pressure calling for detailed discussions on the Bill before its passing, it was referred to a Joint Parliamentary Committee (JPC) constituted for this purpose. It currently remains under review; the JPC is reportedly expected to table its report in the 2021 Winter Session of Parliament. Though the Bill is likely to undergo another round of revisions following the JPC’s review, this is the closest India has come to realising its aspirations of establishing a dedicated and enforceable data protection framework.

This Bill carries forward the choice of a distinct regulatory body, though questions remain on the degree of its independence, given the direct control granted to the central government in appointing its members and funding the DPA.

Conceptualising an Independent DPA

The Srikrishna Committee’s 2017 white paper and its 2018 report on the PDP Bill discuss the need for a regulator in the context of enforcement of its provisions. However, the DPA under the PDP Bill is tasked with extensive powers to frame detailed regulations and codes of conduct to inform the day-to-day obligations of data fiduciaries and processors. To be clear, the standard-setting function for a regulator entails laying down the standards based on which regulated entities (i.e. the data fiduciaries) will be held accountable, and the manner in which they may conduct themselves while undertaking the regulated activity (i.e. personal data processing). This is in addition to its administrative and enforcement, and quasi-judicial functions, as outlined below:

Functions of the DPA under the PDP Bill 2019

At this stage, it is important to note that the choice of regulation via a regulator is distinct from the administration of the Bill by the central or state governments. Creating a distinct regulatory body allows government procedures to be replaced with expert-driven decision-making to ensure sound economic regulation of the sector. At the same time, the independence of the regulatory authority insulates it from political processes. The third advantage of independent regulatory authorities is the scope for ‘operational flexibility’, which is embodied in the relative autonomy of its employees and its decision-making from government scrutiny.

This is also the rationale provided by the Srikrishna Committee in stating their choice to entrust the administration of the data protection law to an independent DPA. The 2017 white paper that preceded the 2018 Srikrishna Committee Report proposed a distinct regulator to provide expert-driven enforcement of laws for the highly specialised data protection sphere. Secondly, the regulator would serve as a single point of contact for entities seeking guidance and will ensure consistency by issuing rules, standards, and guidelines. The Srikrishna Committee Report concretised this idea and proposed a sector-agnostic regulator that is expected to undertake expertise-driven standard-setting, enforcement, and adjudication under the Bill. The PDP Bill carries forward this conception of a DPA, which is distinct from the central government.

Conceptualised as such, the DPA has a completely new set of questions to contend with. Specifically, regulatory bodies require additional safeguards to overcome the legitimacy and accountability questions that arise when law-making is carried out not by elected members of the legislature, but via the unelected executive. The DPA would need to incorporate democratic decision-making processes to overcome the deficit of public participation in an expert-driven body. Thus, the meta-objective of ensuring autonomous, expertise-driven, and legitimate regulation of personal data processing necessitates that the regulator has sufficient independence from political interference, is populated with subject matter experts and competent decision-makers, and further has democratic decision-making procedures.

Further, the standard-setting role of the regulator does not receive sufficient attention in terms of providing distinct procedural or substantive safeguards either in the legislation or public policy guidance.

Reconnaissance under the PDP Bill: How well does it guide the DPA?

At this time, the PDP Bill is the primary guidance document that defines the DPA and its overall structure. India also lacks an overarching statute or binding framework that lays down granular guidance on regulation-making by regulatory agencies.

The PDP Bill, in its current iteration, sets out skeletal provisions to guide the DPA in achieving its objectives. Specifically, the Bill provides guidance limited to the following:

Parliamentary scrutiny of regulations: The DPA must table all its regulations before the Parliament. This is meant to accord legislative scrutiny to binding legal standards promulgated by unelected officials.
Consistency with the Act: All regulations should be consistent with the Act and the rules framed under it. This integrates a standard of administrative law to a limited extent within the regulation-making process.

However, India’s past track record indicates that regulations, once tabled before the Parliament, are rarely questioned or scrutinised. Judicial review is typically based on ‘thin’ procedural considerations such as whether the regulation is unconstitutional, arbitrary, ultra vires, or goes beyond the statutory obligations or jurisdiction of the regulator. In any event, judicial review is possible only when an instrument is challenged by a litigant, and, therefore, it may not always be a robust ex-ante check on the exercise of this power. A third challenge arises where instruments other than regulations are issued by the regulator. These could be circulars, directions, guidelines, and even FAQs, which are rarely bound by even the minimal procedural mandate of being tabled before the Parliament. To be sure, older regulators including the Reserve Bank of India (RBI) and the Securities and Exchange Board of India (SEBI) also face similar issues, which they have attempted to address through various methods including voluntary public consultations, stakeholder meetings, and publication of minutes of meetings. These are useful tools for the DPA to consider as well.

Apart from these, specific guidance is provided with respect to issuing and approving codes of practice and issuing directions as follows:

Codes of practice: The DPA is required to (i) ensure transparency,^{^[1]} (ii) consult with other sectoral regulators and stakeholders, and (iii) follow a procedure to be prescribed by the central government prior to the notification of codes of practice under the Bill.^{^[2]}
Directions: The DPA may issue directions to individual, regulated entities or their classes from time to time, provided these entities have been given the opportunity to be heard by the DPA before such directions are issued.^{^[3]}

However, the meaning of transparency and the process for engaging with sectoral regulators remains unspecified under the Bill. Furthermore, the central government has been provided vast discretion to formulate these procedures, as the Bill does not specify the principles or outcomes sought to be achieved via these procedures. The Bill also does not specify instances where such directions may be issued and in which form.

Thus, as per its last publicly available iteration, the Bill remains silent on the following:

The principles that may guide the DPA in its functioning.
The procedure to be followed for issuing regulations and other subordinate legislation under the Bill.
The relevant regulatory instruments, other than regulations and codes of practice – such as circulars, guidelines, FAQs, etc. – that may be issued by the DPA.
The specifics regarding the members and employees within the DPA who are empowered to make these regulations.

It is unclear whether the JPC will revise the DPA’s structure or recommend statutory guidance for the DPA in executing any of its functions. This is unlikely, given that parent statutes for other regulators typically omit such guidance. As a result, the DPA may be required to make intentional and proactive choices on these matters, much like their regulatory counterparts in India. These are discussed in the section below.

Envisaging a Proactive Role for the DPA

As the primary regulatory body in charge of the enforcement of the forthcoming data protection framework, what should be the role of the DPA in setting standards for data protection?

The complexity of the subject matter, and the DPA’s role as the frontline body to define day-to-day operational standards for data protection for the entire digital economy, necessitates that it develop transparent guiding principles and procedures. Furthermore, given that the DPA’s autonomy and capacity are currently unclear, the DPA will need to make deliberate choices regarding how it conducts itself. In this regard, the skeletal nature of the PDP Bill also allows the DPA to determine its own procedures to carry out its tasks effectively.

This is not uncommon in India: various regulators have devised frameworks to create benchmarks for themselves. The Airports Economic Regulatory Authority (AERA) is obligated to follow a dedicated consultation process as per an explicit transparency mandate under the parent statute. However, the Insolvency and Bankruptcy Board of India (IBBI) has, on its own initiative, formulated regulations to guide its regulation-making functions. In other cases, consultation processes have been integrated into the respective framework through judicial intervention: the Telecom Regulatory Authority of India (TRAI) has been mandated to undertake consultations through judicial interpretation of the requirement for transparency under the Telecom Regulatory Authority of India Act, 1997 (TRAI Act).

In this regard, we develop a list of considerations that the DPA should look to address while carrying out its standard-setting functions. We also draw on best practices by Indian regulators and abroad, which can help identify feasible solutions for an effective DPA for India.

The choice of regulatory instruments

The DPA is empowered to issue regulations, codes of practice, and directions under the Bill. At the same time, regulators in India routinely issue other regulatory instruments to assign obligations and clarify them. Some commonly used regulatory instruments are outlined below. The terms used for instruments are not standard across regulators, and the list and description set out below outline the main concepts and not fixed labels for the instruments.

Overview of regulatory instruments

	Circulars and Master Circulars	Guidelines	FAQs	Directions
Content	Circulars are used to prescribe detailed obligations and prohibitions for regulated entities and can mimic regulations. Master circulars consolidate circulars on a particular topic periodically.	These may be administrative or substantive, depending on the practice of the regulator in question.	Issued in public interest by regulators to clarify the regulatory framework administered by them. They cannot prescribe new standards or create obligations.	Issued to provide focused instructions to individual entities or class of entities in response to an adjudicatory action or in lieu of a current challenge.
Binding character	They are generally binding in the same manner as regulations and rules. However, if they go beyond the parent Act or existing rules and regulations, they may be struck down following a judicial review.	They may or may not be binding depending upon the language employed or the regulator’s practice.	Unclear whether these are binding and to what extent. However, crucial clarifications on important concepts sometimes emerge from FAQs.	Binding in respect of the class of regulated entities to whom this is issued.
Parliamentary scrutiny	Unlike regulations, these do not have to be laid before the Parliament.

Thus, all these instruments, to varying degrees, have been used to create binding obligations for regulated entities. The choice of regulatory instrument is not made systematically. Indeed, even a hierarchy of instruments and their functions are not clearly set out by most regulators. The rationale for deciding why a circular is issued as against a regulation is also unclear. A study on regulatory performance in India by Burman and Zaveri (2018) has highlighted an over-reliance on instruments such as circulars. As per their study, between 2014 and 2016, RBI and SEBI issued 1,016 and 122 circulars, as against 48 and 51 regulations, respectively. These circulars are not bound by the same pre-consultative mandate nor are they mandated to be laid before the Parliament. While circulars may have been intended for routine to routinely used to lay down administrative or procedural requirements, the study narrows its frame of reference to circulars which lay down substantive regulatory requirements. In this instance, it is unclear why parliamentary scrutiny is mandated for regulations alone, and not for instruments like circulars and directions, even though they lay down similarly substantive requirements. Furthermore, there have also been instances where certain instruments like FAQs have gone beyond their advisory scope to provide new directions or definitions that were not previously shared under binding instruments like regulations or circulars.

The DPA has been provided specific powers to issue regulations, codes of practice, and directions. However, the rationale for issuing one instead of the other has been absent from the PDP Bill so far. In such a scenario, it is important that the DPA transparently outlines the types of instruments it wishes to use, whether they are binding or advisory, and the procedure to be followed for issuing each.

Pre-legislative consultative rule-making

Participatory and consultative processes have emerged as core components of democratic rule-making by regulators. Transparent consultative mechanisms could also ameliorate capacity challenges in a new regulator (particularly for technical matters) and help enhance public confidence in the regulator.

In India, several regulators have adopted consultation mechanisms even when there is no specific statutory requirement. SEBI and IBBI routinely issue discussion papers and consultation papers. The RBI also issues draft instruments soliciting comments. As discussed previously, TRAI and AERA have distinct transparency mandates under which they carry out consultations before issuing regulations. However, these processes are not mandated all forms of subordinate legislation. Taking cognizance of this, the Financial Sector Legislative Reform Committee (FSLRC) has recommended transparency in the regulation-making process. This was carried forward by the Financial Stability and Development Council (FSDC), which recommended that consultation processes should be a prerequisite for all subordinate legislations, including circulars, guidelines, etc. A study on regulators’ adherence to these mandates, spanning TRAI, AERA, SEBI, and RBI, demonstrated that this pre-consultation mandate is followed inconsistently, if at all. Predictable consultation practices are therefore critical.

Furthermore, the study stated that it could not determine whether the consultation processes yielded meaningful participation, given that regulators are not obligated to disclose how public feedback was integrated into the rule-making process. Subordinate legislations issued in the form of circulars and guidelines also do not typically undergo the same rigorous consultation processes. Thus, an ideal consultation framework would comprise:

Publication of the draft subordinate legislation along with a detailed explanation of the policy objectives. Further, the regulator should publish the internal or external studies conducted to arrive at the proposed legislation to engender meaningful discussion.
Permitting sufficient time for the public and interested stakeholders to respond to the draft.
Publishing all feedback received for the public to assess, and allowing them to respond to the feedback.

However, beyond specifying the manner of conducting consultations, it will be important for the DPA to determine where they are mandatory and binding, and for which type of subordinate legislations. These are discussed in the next section.

Choice of consultation mandates for distinct regulatory instruments

While the Bill provides for consultation processes for issuing and approving codes of practice, no such mechanism has been set out for other instruments. Nevertheless, specifying consultation mandates for different regulatory instruments is important to ensure that decision-making is consistent and regulation-making remains bound by transparent and accountable processes. As discussed above, regulatory instruments such as circulars and FAQs are not necessarily bound by the same consultation mandates in India. This distinction has been clarified in more sophisticated administrative law frameworks abroad. For instance, under the Administrative Procedures Act in the United States (US), all substantive rules made by regulatory agencies are bound by a consultation process, which requires notice of the proposed rule-making and public feedback. This does not preclude the regulatory agency from issuing clarifications, guidelines, and supplemental information on the rules issued. These documents do not require the consultation process otherwise required for formal rules. However, they cannot be used to expand the scope of the rules, set new legal standards, or have the effect of amending the rules. Nevertheless, agencies are not precluded from choosing to seek public feedback on such documents.

Similarly, the Information Commissioner’s Office in the United Kingdom (UK) takes into consideration public consultations and surveys while issuing toolkits and guidance for regulated entities on how to comply with the data protection framework in the UK.

Here, the DPA may choose to subject strictly binding instruments like regulations and codes of practice to pre-legislative consultation mandates, while softer mechanisms like FAQs may be subject to the publication of a detailed outline of the policy objective or online surveys to invite non-binding, advisory feedback. For each of these, the DPA will nonetheless need to create specific criteria by which it classifies instruments as binding and advisory, and further outline specific pre-legislative mandates for each category.

Framework for issuing regulatory instruments and instructions

While the DPA is likely to issue several instruments, the system based on which these instruments will be issued is not yet clear. Without a clearly thought-out framework, different departments within the regulator typically issue a series of directions, circulars, regulations, and other instruments. This raises questions regarding the consistency between instruments. This also requires stakeholders to go through multiple instruments to find the position of law on a given issue. Older Indian regulators are now facing challenges in adapting their ad hoc system into a framework. For example, the RBI currently issues a series of circulars and guidelines that are periodically consolidated on a subject-matter basis as Master Circulars and Master Directions. These are then updated and published on their website. IBBI also publishes handbooks and information brochures that consolidate instruments in an accessible manner.

While these are useful improvements, these practices cannot keep pace with rapid changes in regulatory instructions and are not complete or user-friendly (for example, the subject-matter based consolidation does not allow for filtering regulatory instructions by entity). Other jurisdictions have developed different techniques such as formal codification processes to consolidate regulations issued by government agencies under one unified code, register, or handbook, websites that allow for searches based on different parameters (subject-matter, type of instrument, chronology, entity-based), and guides tailored to different types of entities. The DPA, as a new regulator, can learn from this experience and adopt a consistent framework right from the beginning.

Further, an ethos of responsive regulation also requires the DPA to evaluate and revise directions and regulations periodically, in response to market and technology trends. A commitment to periodic evaluation of subordinate legislations entrenched in the rules is critical to reducing the dependence on officials and leadership, which may change. For instance, the IBBI has set out a mandatory review of regulations issued by it every three years.

Dedicating capacity for drafting subordinate legislations

The DPA has been granted the discretion to appoint experts and staff its offices with the personnel it needs. A study of European data protection authorities shows that by the time the General Data Protection Regulation, 2016 became effective, most of the authorities increased the number of employees with some even reporting a 240% increase. The annual spending on the authorities also went up for most countries. While these authorities do not necessarily frame subordinate legislations, they nonetheless create guidance toolkits and codes of practice as part of their supervisory functions.

In this regard, the DPA will need to ensure it has dedicated capacity in-house to draft subordinate legislations. Since regulators are generally seen as enforcement authorities, there is inadequate investment in capacity-building for drafting legislations in India.

Moreover, considering the multiplicity of instruments and guidance documents the DPA is expected to issue, it may seek to create templates for these instruments, along with compulsory constituents of different types of instruments. For instance, the Office of the Australian Information Commissioner is required to include a mandatory set of components while issuing or approving binding industry codes of practice.

Conclusion

The Personal Data Protection Bill, 2019 (in the final form recommended by the JPC and accepted by the MeitY) will usher in a new chapter in India’s data protection timeline. While the Bill will finally effectuate a nearly comprehensive data protection framework for India, it will also establish a new regulatory framework that sets up a new regulator, the DPA, to oversee the new data protection law. This DPA will be empowered to regulate entities across sectors and is likely to determine the success of the data protection law in India.

Furthermore, the DPA must not only contend with the complexity of markets and the fast pace of technological change, but it must also address anticipated regulatory capacity deficits, low levels of user literacy, the number and diversity of enities within its regulatory ambit, and the need to secure individual privacy within and outside the digital realm.

Thus, looking ahead, we must account for the questions of governance that the forthcoming DPA is likely to face, as these will directly impact how entities and citizens engage with the DPA. In India, regulatory agencies adopt distinct choices to fulfil their functions. Regulators have also fared variably in ensuring transparent and accountable decision-making driven by demonstrable expertise. Even if the final form of the PDP Bill does not address these gaps, the DPA has the opportunity to integrate benchmarks and best practices as discussed above within its own governance framework from the get-go as it takes on its daunting responsibilities under the PDP Bill.

(The authors are Research Fellow, Law, Technology and Society Initiative and Project Lead, Regulatory Governance Project respectively at the National Law School of India University, Bangalore. Views are personal.)

This post was reviewed by Vipul Kharbanda and Shweta Mohandas

References

For a discussion on distinct regulatory choices, please see TV Somanathan, The Administrative and Regulatory State in Sujit Choudhary, Madhav Khosla, et al. (eds), Oxford Handbook of the Indian Constitution (2016).
On best practices for consultative law-making, see generally European Union Better Regulation Communication, Guidelines for Effective Regulatory Consultations (Canada), and OECD Best Practice Principles for Regulatory Policy: The Governance of Regulators, 2014.

^{^[1]} Personal Data Protection Bill 2019, § 50(3).

^{^[2]} Personal Data Protection Bill 2019, § 50(4).

^{^[3]} Personal Data Protection Bill 2019, § 51.

For more details visit https://cis-india.org/internet-governance/blog/trishi-jindal-and-s-vivek-beyond-the-pdp-bill

Big Tech’s privacy promise to consumers could be good news — and also bad news

Rajat Kathuria and Isha Suri — 2023-01-18T23:25:28Z

Rajat Kathuria, Isha Suri write: Its use as a tool for market development must balance consumer protection, innovation, and competition.

In February, Facebook, rebranded as Meta, stated that its revenue in 2022 is anticipated to reduce by $10 billion due to steps undertaken by Apple to enhance user privacy on its mobile operating system. More specifically, Meta attributed this loss to a new AppTrackingTransparency feature that requires apps to request permission from users before tracking them across other apps and websites or sharing their information with and from third parties. Through this change, Apple effectively shut the door on “permissionless” internet tracking and has given consumers more control over how their data is used. Meta alleged that this would hurt small businesses benefiting from access to targeted advertising services and charged Apple with abusing its market power by using its app store to disadvantage competitors under the garb of enhancing user privacy.

Access the full article published in the Indian Express on April 13, 2022

For more details visit https://cis-india.org/internet-governance/blog/indian-express-rajat-kathuria-isha-suri-big-tech-consumers-privacy-policy

Inputs to the Report on the Non-Personal Data Governance Framework

sumandro — 2020-12-30T09:40:52Z

This submission presents a response by researchers at the Centre for Internet and Society, India (CIS) to the draft Report on Non-Personal Data Governance Framework prepared by the Committee of Experts under the Chairmanship of Shri Kris Gopalakrishnan. The inputs are authored by Aayush Rathi, Aman Nair, Ambika Tandon, Pallavi Bedi, Sapni Krishna, and Shweta Mohandas (in alphabetical order), and reviewed by Sumandro Chattapadhyay.

Text of submitted inputs: Read (PDF)

Report by the Committee of Experts on Non-Personal Data Governance Framework: Read (PDF)

Inputs

Clause 3.7 (v): The role of the Indian government in the operation of data markets

While highlighting the potential for India to be one of the top consumer and data markets of the world, it also sheds light on the concern about the possibility of data monopolies. The clause envisions the role of the Indian government as a regulator and a catalyst for domestic data markets.

In doing so, the clause does not acknowledge that the proactive and dominant roles of the Indian government in generation and reuse of data, based on the existing data collection practices, as well as the provisions that have been given, as under the compulsory sharing provisions in the Report, and would continue to be given by the Personal Data Protection Bill. In reality, the Indian government’s role is not just of a catalyst but also of a key player, potentially with monopolistic market power, in the domestic data market, especially due to the ongoing data marketplace initiatives as detailed in published policy and vision documents. [1]

Clause 3.8 (iv): Introducing collective privacy

The introduction of collective privacy has initiated an overdue discussion at the policy level to arrive at privacy formulations that account for limitations in the contemporary dominant social, legal and ethical paradigms of privacy premised on individual interests and personal harm. The notion of collective privacy has garnered contemporary attention with the rise of data processing technologies and business models that thrive on the collection and processing of aggregate information.

While the Report acknowledges that collective privacy is an evolving concept, it doesn’t attempt to define either collective or what privacy could entail in the context of a collective. The postulation of collective privacy as a legally binding right is bereft with challenges in both domestic and international legal frameworks. [2]

Central to these challenges is the representation of the group of the entity. While the Report illustrates harms that may be incurred by certain collectives that collective privacy could protect against, these illustrated collectives are already recognised in law as rights-holding groups (society members, for example), and/or share pre-determined attributes (sexual orientation, for example).

The Report does not acknowledge that the very technological processes that may have rendered the articulation of collective privacy necessary, also are intended to create ad-hoc and newer sets of individuals or groups with shared attributes. [3] In doing so, the Report furthers an ontology of groups having intuitive, predetermined attributes that exist naturally, or in law, whereas the intervention of data collection and processing technologies can determine shared group attributes afresh. Moreover, the Report also ignores that predetermined attributes are static, and in doing so, ignores a vast existing literature speaking to fluidity of identities and the intersectionality of identities that individuals in groups occupy. [4] We fully appreciate the challenges these pose in the determination of the legal contours of collective privacy. Much of the Report’s recommendations are premised on the idea of a predetermined collective, rendering more granular exploration of these ideas urgent.

Further, the Report also puts forth a limited conception of privacy as a safeguard against data-related harms that may be caused to collectives. In doing so, it dilutes the conceptualisation of individual privacy as articulated in Justice K. S. Puttaswamy (Retd.) and Anr. vs Union Of India And Ors. Notwithstanding this dilution, the illustrations also only indicate harms that may be caused by private actors. Any further recommendations should envision the harms that may also be caused by public data-driven processes, such as those incubated within the state machinery.

Clause 4.1 (iii) and Recommendation 1: Defining Non-Personal Data

The Report proposes the definition of non-personal data to include (i) data that was never related to an identified or identifiable natural person, and (ii) aggregated, anonymised personal data such that individual events are “no longer identifiable”. In doing so, they have attempted to extend protections to categories of data that fall outside the ambit of the Personal Data Protection Bill, 2019 (hereafter “PDP Bill”). The Report is cognizant of the fallible nature of anonymization techniques but fails to indicate how these may be addressed. The test of anonymization in regarding data as non-personal data requires further clarification. Anonymization, in and of itself, is an ambiguous standard. Scholarship has indicated that anonymised data may never be completely anonymous. [5] Despite this, the PDP Bill proposes a high threshold of zero-risk of anonymization in relation to personal data, to mean “such irreversible process of transforming or converting personal data to a form in which a data principal cannot be identified”. From a plain reading, it appears that the Report proposes a lower threshold of the anonymization requirements governing non-personal data. It is unclear how non-personal data would then be different from inferred data as described within the definition of personal data under the PDP Bill. This adds regulatory uncertainty making it imperative for the Committee to articulate bright-line, risk-based principles and rules for the test of anonymization. Such rules should also indicate the factors that ought to be taken into account to determine whether anonymization has occurred and the timescale of reference for anonymization outcomes. [6]

The recommendation also states that the data principal should "also provide consent for anonymisation and usage of this anonymized data while providing consent for collection and usage of his/her personal data". However the framing of this recommendation fails to mention the responsibility of the data fiduciary to provide notice to the data principal about the usage of the anonymized data while seeking the data principal’s consent for anonymization. The notice provided to the data principal should provide clear indication that consent of the data principal is based on their knowledge of the use of the anonymized data.

Clause 4.8 (i), (ii): Function of data custodians

The Report does not make it clear who may perform the role of data custodians. The use of data fiduciary indicates the potential import of the definition of ‘data fiduciary’ as specified under Clause 3.13 of the PDP Bill. However, this needs to be further clarified.

Clause 4.8 (iii): Data custodians’ “duty of care”

As is outlined in the following section on data trustees, it can be difficult for a singular entity to maintain a duty of care and undertake actions with the best interest of a community when that community consists of sub-communities that may be marginalised. Further, ‘duty of care’, ‘best interest’, and ‘absence of harm’ are not sufficient standards for data processing by data custodians. Recommendations to the effect of obligating data custodians to uphold the rights of data principals, including economic and fundamental rights need to be incorporated in the framework.

Clause 4.9: Data trustees

The committee’s suggestion that the “most appropriate representative body” should be the data trustee—that often being either the corresponding government entity or community body— is reasonable at face value. However, in the absence of any clear principles defining what constitutes “most appropriate” there are a number of potential issues that can appear:

Lack of means for selecting a data trustee: The report makes note of the fact that both private and public entities can be selected to be data trustees but offers no principles on how these data trustees can be selected, i.e. whether they are to be directly selected by the members of a community, and if so how. Any selection criteria or process prescribed has to keep in mind the following point regarding the potential lack of representation for marginalised communities that could arise from a direct selection of a data trustee by a group of people.

Issues of having a single data trustee for large scale communities and when dealing with marginalised communities: The report assumes that in instances wherein a community is spread across a geographic region, or consists of multiple sub-communities, then the data trustee will be the closest shared government authority (for example, the Ministry of Health and Family Welfare, Government of India being the data trustee for data regarding diabetes among Indian citizens).

This idea of a singular data trustee assumes that the ‘best interests’ of a community are uniform across that community. This can prove problematic especially when dealing with data obtained from marginalised communities that forms a part of a wider dataset. It is entirely possible to imagine that a smaller disenfranchised community may have interests that are not aligned with the general majority. In such a situation the Report is unclear as to whether the data trustee would have to ensure that the best interests of all groups are maintained, or would they be responsible for ensuring the best interests of the largest number of people within that community. There are power differentials between citizens, government agencies, and other entities described by the Report. This places citizens at risk of abuse of power by government entities in their role as trustees, who are effectively being empowered through this policy framework as opposed to a representative mechanism. It is recommended that data trustees be appointed by relevant communities through clear and representative mechanisms. Additionally, any individual should be able to file complaints regarding the discharge of community trust by data trustees. This is necessary as any subsequent rights vested in the community can only be exercised through the data trustee, and become unenforceable in the lack of an appropriate data trustee.

Any legislation that arises on the basis of this report will therefore have to not only provide a means for selecting the data trustee, but also safeguards for ensuring that data collected from marginalised communities are used keeping in mind their specific best interests—with these best interests being informed through consultation with that community.

Clause 4.10 (iii): Data trusts

Section 4.10 (iii) notes that data custodians may voluntarily share data in these data trusts. However it is unclear if such sharing must be done with the express consent of the relevant data trustee.

Clause 4.10 (iv): Mandatory sharing and competition

The fundamental premise of a mandatory data sharing regime seems increasingly distant from its practical impacts. The EU which earlier championed the cause now seems reluctant to further it on the face of studies which skews towards counteractive impacts of such steps. Such steps could apply to huge volumes of first-party data companies collect on their own assets, products and services, even though such data are among the least likely to create barriers to entry or contribute to abuses of dominant positions. [7] This is hence likely to bring in more chilling effect on innovation and investment than a pro-competition environment. The velocity of big data also adds to the futility of such data sharing mandates. [8] It is recommended that a sectoral analysis of this mandate be undertaken instead of an overarching stipulation.

The Report suggests extensive data sharing without addressing the extent of obligation on the private players to submit to these requests and process them. The availability of meta-data about the data collected may be made easily accessible under mandates of transparency. However, the access to the detailed underlying data will be difficult in most cases due to the current structure of entities functioning in cyberspace, evidenced by the lack of compliance to such mandates by Courts of Law in the EU. Such a system can easily eliminate the comparative advantage of smaller players, helping larger players with more money at their disposal enabling their growth and throttling the smaller players. It could have serious implications on data quality and integrity through the sharing of erroneous data. Access to superior quality digital services in India may also have to be compromised. If this regime is furthered without amends to address these concerns, it might end up counter productive.

Clause 5.1 (iv): Grievance redressal against state’s role

This clause acknowledges the vast potential for government authorities and other bodies to abuse their power as data trustee. In addition, it should describe the setting up of impartial and accessible mechanisms for citizens to complain against such abuse of power and appropriate penalties, including the removal of the data trustee.

Chapter 7, Recommendation 5: Purpose of data-sharing

Recommendation 5 leaves scope for “national security” as a sovereign purpose for data sharing. This continues to be in line with the trend of having an overarching national security clause, as in the Personal Data Protection Bill, 2019. There could be provisions made to enable access to data for sovereign purposes without such broad definition, replacing it based on constitutional terms which will limit it to the confines laid down in the Constitution. This will effectively curb any misuse of the provision and strongly embed the proposed regulation of non-personal data on constitutional ethos. This can also prevent future conflicts with the fundamental rights.

Platform companies have leveraged their position in society to take on an ever-greater number of quasi-public functions, exercising new forms of unaccountable, transnational authority. It is not difficult to imagine that this trend can continue to non-platform companies, or even taken forward by these very entities which also have access to a large chunk of non-personal data. A strict division between sovereign purposes and core public interest purposes seems difficult. However, it is imperative to have a clearer definition of core public interest purposes and sovereign purposes. The broad based definition may facilitate reduced accountability. Separating government actions from sovereign purposes could bring forth the power imbalance between the State and its people, while in the case of the non-governmental entities, it will facilitate encroachment of government functions by private players. Both these cases may not consider the best interest of the data generators, or the people at large.

Clause 7.1 (i): Data needs of law enforcement

Clause 7.1 (i) allows for acquisition of data governed by this framework for crime mapping, devising anticipation and preventive measures, and for investigations and law enforcement. While this may be necessary to be granted to law enforcement in certain cases, this should happen only with an express permission of a court of law. Blanket executive access allows higher possibility of misuse by the people involved in law enforcement.

Clause 7.2 (iv): Use of health data as a pilot

The clause suggests the use of health sector data as a pilot use-case. This is highly undesirable due to the inherent nature of high sensitivity of the larger part of data related to the health sector. The high vulnerability of such data to harm the data principals should act as a deterrent in using this as the pilot use-case. Given the mass availability of data related to the health sector due to the pandemic, it creates further points of vulnerabilities which can be illegally monetised and misappropriated. It is recommended that this proposal be scrapped altogether.

Clause 7.2 (iii): Power of government bodies

As per this clause, data trustees or government bodies (who could also be acting as data trustees) can make requests for data sharing and place such data in appropriate data infrastructures or trusts. This presents a conflict of interest, as a data trust or government body can empower itself to be the data trustee. Such cases should be addressed within the scope of the framework.

Clause 8.2 (vii): Level-playing field for all Indian actors

In terms of this clause the “Non-Personal Data Authority (Authority) will ensure a level playing field for all Indian actors to fulfil the objective of maximising Indian data’s value to the Indian economy”. The emphasis on ensuring a level playing field for only Indian actors instead of non-discriminatory platform for all concerned actors irrespective of the country/nationality of the actor has the potential of violating India’s trade obligations under the WTO. Member states of the WTO are essentially restricted from discriminating between products and services coming from different WTO Members, and between foreign and domestic products and services unless they can avail of exceptions. There is also no clarity on what constitutes ‘Indian Actors’, would a Multi-National Corporation with its headquarters in a foreign State, but its subsidiaries in India also come within its ambit.

Clause 8.2 (x): Composition of the Authority

Clause 8.2 (x) states that the Authority will have some members with relevant industry experience. However, apart from this clause, the report is silent on the composition of the Authority. The report recognises that Authority will need individuals/organisations with specialised knowledge, i.e. data governance, technology, latest research and innovation in the field of non-personal data), however, it does not mention or refer to the role of civil society organisations and the need for representation from such organisations in the Authority.

The report frequently alludes to non-personal data being used for the best interest of the data principal and therefore, it is essential that the composition of the Authority reflect the inherent asymmetry of power between the data principal and the State. Considering that the Authority will also be responsible for sharing of community data and with determining the code of conduct for sharing of such data, it is important that the Authority also has adequate representation from civil society organisations along with groups or individuals having the necessary technological and legal skills.

Clause 8.2 (iii) and (vi): Roles and Responsibility of the Authority

A majority of the datasets in the country comprise of ‘mixed datasets’, i.e. it consists of both personal and non-personal data. However, there is lack of clarity about the coordination between the Data Protection Authority constituted under the PDP Bill and the Non-Personal Data Authority with regard to the regulation of such datasets. The Report refers to the European Union which provides that the Non-Personal Data Regulation applies to the Non-Personal Data of mixed datasets; if the Non-Personal Data part and the personal data parts are ‘inextricably linked’, the General Data Protection Regulation apply to the whole mixed dataset. However, it is unclear whether the Report also proposes the same mechanism for the regulation of mixed datasets.

Further, the contours of the enforcement role of the Committee should be specified and clearly laid down. Will the Committee also have penal powers as prescribed for the Data Protection Authority under the PDP Bill? Also, will the privacy concerns emanating from the risk of re-anonymisation of data be addressed by the NPD Committee or by the DPA under the PDP Bill. Ideally, it should be specified that any such privacy concerns will fall within the domain of the DPA as the data is then converted into personal data and the DPA will be empowered to deal with such issues.

Endnotes

[1] See Ministry of Health and Family Welfare. (2020). National Digital Health Blueprint. Government of India. https://main.mohfw.gov.in/sites/default/files/Final%20NDHB%20report_0.pdf; Tandon, A. (2019). Big Data and Reproductive Health in India: A Case Study of the Mother and Child Tracking System. https://cis-india.org/raw/big-data-reproductive-health-india-mcts

[2] Taylor, L., Floridi, L., van der Sloot, B. eds. (2017) Group Privacy: new challenges of data technologies. Dordrecht: Springer.

[3] Mittelstadt, B. (2017). From Individual to Group Privacy in Big Data Analytics. Philos. Technol. 30, 475–494.

[4] See Taylor, L., Floridi, L., van der Sloot, B. eds. (2017) Group Privacy: new challenges of data technologies. Dordrecht: Springer; Tisne, M. (n.d). The Data Delusion: Protecting Individual Data Isn't Enough When The Harm is Collective. Stanford Cyber Policy Centre. https://cyber.fsi.stanford.edu/publication/data-delusion

[5] Rocher, L., Hendrickx, J.M. & de Montjoye, Y. (2019). Estimating the success of re-identifications in incomplete datasets using generative models. Nat Commun 10, 3069 . https://doi.org/10.1038/s41467-019-10933-3

[6] Finck, M. & Pallas, F. (2020). They who must not be identified—distinguishing personal from non-personal data under the GDPR. International Data Privacy Law, 10 (1), 11–36. https://doi.org/10.1093/idpl/ipz026

[7] European Commission (2020). Communication From The Commission To The European Parliament, The Council, The European Economic And Social Committee And The Committee Of The Regions: A European strategy for data. https://eur-lex.europa.eu/legal-content/EN/TXT/?qid=1593073685620&uri=CELEX:52020DC0066

[8] Modrall, Jay. (2019). Antitrust risks and Big Data. Norton Rose Fullbright. https://www.nortonrosefulbright.com/en-in/knowledge/publications/64c13505/antitrust-risks-and-big-data

For more details visit https://cis-india.org/raw/inputs-to-report-on-non-personal-data-governance-framework

The PDP Bill 2019 Through the Lens of Privacy by Design

2020-11-13T07:51:03Z

This paper evaluates the PDP Bill based on the Privacy by Design approach. It examines the implications of Bill in terms of the data ecosystem it may lead to, and the visual interface design in digital platforms. This paper focuses on the notice and consent communication suggested by the Bill, and the role and accountability of design in its interpretation.

Background

The Personal Data Protection (PDP) Bill, 2019 was introduced in the Lok Sabha on December 11, 2019 by the Minister of Electronics and Information Technology. The Bill aims to provide for protection of personal data of individuals, and establishes a Data Protection Authority for the same [1]. The PDP Bill, 2019 contains several clauses that have implications on the visual design of digital products. These include the specific requirements for communication of notice and consent at various stages of the product. The Bill also introduces the Privacy by Design policy. Privacy by Design (PbD), as a concept, was proposed by Ann Cavoukian in the 1990s, with the purpose of approaching privacy from a design-thinking perspective [2]. She describes this perspective to be holistic, interdisciplinary, integrative, and innovative. The approach suggests that privacy must be incorporated into networked data systems and technologies, by default [3]. It challenges the practice of enhancing privacy as an afterthought. It expects privacy to be a default setting, and a proactive (not reactive) measure that would be embedded into a design in its initial stage and throughout the life cycle of the product [4]. While PbD is a conceptual framework, it’s application can change the way digital platforms are created and the way in which people interact with them. From devising a business model, to making technological decisions, PbD principles can make privacy integral to the processes and standards of a digital platform.

The PDP Bill states that data fiduciaries are required to prepare a Privacy by Design policy and have it certified by the Data Protection Authority. According to the Bill, the policy would contain the managerial, organisational, business practices and technical systems designed to anticipate, identify and avoid harm to the data principal [5]. It would mention if the technology used in the processing of personal data is in accordance with the certified standards. It would also comprise of the ways in which privacy is being protected throughout the stages of processing of personal data, and that the interest of the individual is accounted for in each of these stages. Once certified by the Data Protection Authority, the data fiduciaries are also required to publish this policy on their website [6]. This forces the data fiduciaries to envision privacy as a fundamental requirement and not an afterthought. Such a policy would have a huge impact in the way digital platforms are conceptualised, both from the technological and the design point of view. The adoption of this policy by digital platforms would enable people to know if their privacy is protected by the companies, and what are the various steps being taken for this purpose. Besides the explicit Privacy by Design policy, the PDP Bill, 2019, also recommends the regulations for data minimisation, establishment of the Data Protection Authority (DPA), and the development of a consent framework. These steps are also part of the Privacy by Design approach.

This paper evaluates the PDP Bill based on the Privacy by Design approach. The Bill’s scope includes both the conceptual and technological aspects of a digital platform, as well as the interface aspect that the individual using the platform faces. The paper will hence analyse how PbD approach is reflected in both these aspects. At the conceptual level, it will look at the data ecosystem that the Bill unwittingly creates, and at the interface level, it will critically analyse the Bill’s implication on the notice and consent communication in the digital products. This includes the several points of communication or touchpoints between a company and an individual using their service, as dictated by the Bill, and how they would translate into visual design. Visual design forms an integral part of digital platforms. It is the way in which the platforms interact with the individuals. The choices made by individuals are largely driven by the visual structuring and presentation of information on these platforms. Presently, the interface design in several platforms is being used to perpetuate unethical data practices in the form of dark patterns. Dark Patterns are deceptive user interface interactions, designed to mislead or trick users to make them do something they don’t want to do [7]. The design of the notice and consent touchpoints can significantly influence the enforcement of this Bill, and how it benefits individuals. Moreover, digital platforms may technically follow the regulations but can still be manipulative through their interface design. Thus, the role and accountability of design becomes crucial in the interpretation of the data protection regulations.

The full paper can be read here.

[1] https://prsindia.org/billtrack/personal-data-protection-bill-2019

[2] https://iab.org/wp-content/IAB-uploads/2011/03/fred_carter.pdf

[3] https://iab.org/wp-content/IAB-uploads/2011/03/fred_carter.pdf

[4] https://www.smashingmagazine.com/2019/04/privacy-ux-aware-design-framework/

[5] http://164.100.47.4/BillsTexts/LSBillTexts/Asintroduced/373_2019_LS_Eng.pdf

[6] https://sflc.in/key-changes-personal-data-protection-bill-2019-srikrishna-committee-draft

[7] https://uxdesign.cc/dark-patterns-in-ux-design-7009a83b233c

For more details visit https://cis-india.org/internet-governance/blog/the-pdp-bill-2019-through-the-lens-of-privacy-by-design

Fundamental Right to Privacy — Three Years of the Puttaswamy Judgment

pranav — 2020-08-24T07:46:10Z

Today marks three years since the Supreme Court of India recognised the fundamental right to privacy, but the ideals laid down in the Puttaswamy Judgment are far from being completely realized. Through our research, we invite you to better understand the judgment and its implications, and take stock of recent issues pertaining to privacy.

Amber Sinha dissects the Puttaswamy Judgment through an analysis of the sources, scope and structure of the right, and its possible limitations. [link]

Through a visual guide to the fundamental right to privacy, Amber Sinha and Pooja Saxena trace how courts in India have viewed the right to privacy since Independence, explain how key legal questions were resolved in the Puttaswamy Judgement, and provide an account of the four dimensions of privacy — space, body, information and choice — recognized by the Supreme Court. [link]

Based on publicly available submissions, press statements, and other media reports, Arindrajit Basu and Amber Sinha track the political evolution of the data protection ecosystem in India, on EPW Engage. They discuss how this has, and will continue to impact legislative and policy developments. [link]

For the AI Policy Exchange, Arindrajit Basu and Siddharth Sonkar examine the Automated Facial Recognition Systems (AFRS), and define the key legal and policy questions related to privacy concerns around the adoption of AFRS by governments around the world. [link]

Over the past decade, reproductive health programmes in India have been digitising extensive data about pregnant women. In partnership with Privacy International, we studied the Mother and Child Tracking system (MCTS), and Ambika Tandon presents the impact on the privacy of mothers and children in the country. [link]

While the right to privacy can be used to protect oneself from state surveillance, Mira Swaminathan and Shubhika Saluja write about the equally crucial problem of lateral surveillance — surveillance that happens between individuals, and within neighbourhoods, and communities — with a focus on this issue during the COVID-19 crisis. [link]

Finally, take a dive into the archives of the Centre for Internet and Society to read our work, which was cited in the Puttaswamy judgment — essays by Ashna Ashesh, Vidushi Marda and Bhairav Acharya that displaced the notion that privacy is inherently a Western concept, by attempting to locate the constructs of privacy in Classical Hindu [link], and Islamic Laws [link]; and Acharya’s article in the Economic and Political Weekly, which highlighted the need for privacy jurisprudence to reflect theoretical clarity, and be sensitive to unique Indian contexts [link].

For more details visit https://cis-india.org/internet-governance/blog/fundamental-right-to-privacy-three-years-of-the-puttaswamy-judgment

A Compilation of Research on the PDP Bill

pranav — 2020-03-05T08:04:24Z

The most recent step in India’s initiative to create an effective and comprehensive Data Protection regime was the call for comments to the Personal Data Protection Bill, 2019, which closed last month. Leading up to the comments, CIS has published numerous research pieces with the goal of providing a comprehensive overview of how this legislation would place India within the global scheme, and how the local situation has developed, as well as analysing its impacts on citizens’ rights.

In addition to general and clause-by-clause comments and recommendations, we have compiled an annotated version of the Personal Data Protection Bill, which lays out our commentary in an easy-to-follow format.

Below, you can find our other recent research on Data Protection:

Pallavi Bedi has put together a note on the Divergence between EU’s General Data Protection Regulation (GDPR) and the Personal Data Protection Bill.

In addition, Pallavi has also contrasted the Personal Data Protection Bill with the GDPR and California Consumer Protection Act, in the contexts of jurisdiction and scope, rights of the data principal, obligations of data fiduciaries, exemptions, data protection authority, and breach of personal data.

On IAPP’s blog Privacy Perspectives, D. Shweta Reddy has assessed whether the Personal Data Protection Bill 2019 is sufficient for India to receive adequacy status from the EU.

Along with Justin Sherman, Arindrajit Basu has outlined the key global takeaways from the Personal Data Protection Bill 2019 on Lawfare.

On The Diplomat, Arindrajit has also traced the narrowing localization provisions in India, as well as Vietnam and Indonesia, and studied the actors and geopolitical tussle that has shaped these provisions.

Through a string of publicly available submissions, press statements, and other media reports, Arindrajit and Amber Sinha have tracked the political evolution of the data protection ecosystem in India, and how this has, and will continue to impact legislative and policy developments on EPW Engage.

Gurshabad Grover and Tanaya Rajwade have written on The Wire about how the Personal Data Protection Bill regulates social media.

Amber was also a guest on Suno India’s Cyber Democracy podcast, with Srinivas Kodali, to discuss how the latest version of the Personal Data Protection Bill will impact the right to privacy.

For more details visit https://cis-india.org/internet-governance/blog/compilation-of-research-on-data-protection

Divergence between the General Data Protection Regulation and the Personal Data Protection Bill, 2019

Pallavi Bedi — 2020-02-21T11:08:50Z

Our note on the divergence between the General Data Protection Regulation and the Personal Data Protection Bill can be downloaded as a PDF here.

The European Union’s General Data Protection Regulation (GDPR), replacing the 1995 EU Data Protection Directive came into effect in May 2018. It harmonises the data protection regulations across the European Union. In India, the Ministry of Electronics and Information Technology had constituted a Committee of Experts (chaired by Justice Srikrishna) to frame recommendations for a data protection framework in India. The Committee submitted its report and a draft Personal Data Protection Bill in July 2018 (2018 Bill). Public comments were sought on the bill till October 2018. The Central Government revised the Bill and introduced the revised version of the Personal Data Protection Bill (PDP Bill) on December 11, 2019 in the Lok Sabha.

The PDP Bill has incorporated certain aspects of the GDPR, such as requirements for notice to be given to the data principal, consent for processing of data, establishment of a data protection authority, etc. However, there are some differences and in this note we have highlighted the areas of divergence between the two. It only includes provisions which are common to the GDPR and the PDP Bill. It does not include the provisions on (i) Appellate Tribunal, (ii) Finance, Account and Audit; and (iii) Non- Personal Data.

For more details visit https://cis-india.org/internet-governance/blog/divergence-between-the-general-data-protection-regulation-and-the-personal-data-protection-bill-2019

Announcing Selected Researchers: Welfare, Gender, and Surveillance

sumandro — 2020-02-13T15:04:24Z

We published a Call for Researchers on January 10, 2020, to invite applications from researchers interested in writing a narrative essay that interrogates the modes of surveillance that people of LGBTHIAQ+ and gender non-conforming identities and sexual orientations are put under as they seek sexual and reproductive health (SRH) services in India. We received 29 applications from over 10 locations in India in response to the call, and are truly overwhelmed by and grateful for this interest and support. We eventually selected applications by 3 researchers that we felt aligned best with the specific objectives of the project. Please find below brief profile notes of the selected researchers.

Call for Researchers: URL

Kaushal Bodwal

Kaushal is persuing his MPhil in Sociology at Delhi School of Economics, University of Delhi. He completed his Master's in Sociology at Centre for the Study of Social Systems, Jawaharlal Nehru University after getting a BSc honors degree in Biomedical Sciences from Delhi University. He is one of the founding members of Hasratein: a queer collective, New Delhi. He has been an active spokesperson for Queer and Trans Rights in India and have been on a number of panel discussion on Trans Act 2019 in various campuses. He has also delivered a lecture series on Colonialism and Medicine in Ambedkar University, Kashmiri Gate, Delhi. His areas of interest are Sociology of medicine, gender and medicine, sexuality, religion and biomedical science, intersex studies.

Queerness as disease – a continuing narrative in 21st century India, Kafila, 27 August 2019

What it means to be queer under a regime bent on remaking India on its own ideological terms, Firstpost, 17 January 2020

Rosamma Thomas

Rosamma has worked both as a reporter and as an editor of news reports with newspapers. She currently writes reports for NGOs while also undertaking freelance reporting assignments. She is based in Pune.

India's mining state steps up fight to rein in killer silicosis, The Times of India, 29 June 2016

Doctor may have found early marker for silicosis, but who will fund him?, Newsclick, 18 July 2019

Asbestos poisoning: Raghunath Manwar’s fight for safer work conditions, Newsclick, 9 January 2020

Shreya Ila Anasuya

Shreya is a writer, editor, journalist and performance artist currently based in Calcutta. Her fiction explores the places where myth, memory, history and the performing arts meet. As a journalist, her work explores gender, sexuality, politics, culture and history. She has been published in The Wire, Caravan, Scroll, Mint Lounge, Deep Dives, GenderIT, Helter Skelter, and many more. She is the editor of the digital publication Skin Stories, housed at the non-profit Point of View. She is the writer and narrator of ‘Gul - a story in text, song and dance’ which has been performed in several cities in India. She was a Felix Scholar at SOAS, University of London, from where she has an MA in Anthropology. For a full portfolio, please click here or visit her website.

This project is led by Ambika Tandon, Aayush Rathi, and Sumandro Chattapadhyay at the Centre for Internet and Society, and is supported by a grant from Privacy International.

For more details visit https://cis-india.org/raw/announcing-selected-researchers-welfare-gender-and-surveillance

Comments to the Personal Data Protection Bill 2019

Amber Sinha, Elonnai Hickok, Pallavi Bedi, Shweta Mohandas, Tanaya Rajwade — 2020-02-21T10:13:35Z

The Personal Data Protection Bill, 2019 was introduced in the Lok Sabha on December 11, 2019.

Please view our general comments below, or download as PDF here.

Our comments and recommendations can be downloaded as PDF here.

We have also prepared an annotated version of the Bill, where our detailed comments and recommendations can be viewed alongside the Bill, available as PDF here.

General Comments

1. Executive notification cannot abrogate fundamental rights

In 2017, the Supreme Court in K.S. Puttaswamy v Union of India [1] held the right to privacy to be a fundamental right. While this right is subject to reasonable restrictions, the restrictions have to meet a three fold requirement, namely (i) existence of a law; (ii) legitimate state aim; (iii) proportionality.Under the 2018 Bill, the exemption to government agencies for processing of personal data from the provisions of the Bill in the ‘interest of the security of the State’ [2] was subject to a law being passed by Parliament. However, under Clause 35 of the present Bill, the Central Government is merely required to pass a written order exempting the government agency from the provisions of the Bill.Any restriction on the right to privacy will have to comply with the conditions prescribed in Puttaswamy I. An executive order issued by the central government authorising any agency of the government to process personal data does not satisfy the first requirement laid down by the Supreme Court in Puttaswamy I — as it is not a law passed by Parliament. The Supreme Court while deciding upon the validity of Aadhar in K.S. Puttaswamy v Union of India [3] noted that “an executive notification does not satisfy the requirement of a valid law contemplated under Puttaswamy. A valid law in this case would mean a law passed by Parliament, which is just, fair and reasonable. Any encroachment upon the fundamental right cannot be sustained by an executive notification.”

2. Exemptions under Clause 35 do not comply with the legitimacy and proportionality test

The lead judgement in Puttaswamy I while formulating the three fold test held that the restraint on privacy emanate from the procedural and content based mandate of Article 21 [4]. The Supreme Court in Maneka Gandhi v Union India [5] had clearly established that “mere prescription of some kind of procedure cannot ever meet the mandate of Article 21. The procedure prescribed by law has to be fair, just and reasonable, not fanciful, oppressive and arbitrary” [6]. The existence of a law is the first requirement; the second requirement is that of ‘legitimate state aim’. As per the lead judgement this requirement ensures that “the nature and content of the law which imposes the restriction falls within the zone of reasonableness mandated by Article 14, which is a guarantee against arbitrary state action” [7]. It is established that for a provision which confers upon the executive or administrative authority discretionary powers to be regarded as non-arbitrary, the provision should lay down clear and specific guidelines for the executive to exercise the power [8]. The third test to be complied with is that the restriction should be ‘proportionate,’ i.e. the means that are adopted by the legislature are proportional to the object and needs sought to be fulfilled by the law. The Supreme Court in Modern Dental College & Research Centre v State of Madhya Pradesh [9] specified the components of proportionality standards —

A measure restricting a right must have a legitimate goal;
It must be a suitable means of furthering this goal;
There must not be any less restrictive, but equally effective alternative; and
The measure must not have any disproportionate impact on the right holder

Clause 35 provides extensive grounds for the Central Government to exempt any agency from the requirements of the bill but does not specify the procedure to be followed by the agency while processing personal data under this provision. It merely states that the ‘procedure, safeguards and oversight mechanism to be followed’ will be prescribed in the rules.The wide powers conferred on the central government without clearly specifying the procedure may be contrary to the three fold test laid down in Puttaswamy I, as it is difficult to ascertain whether a legitimate or proportionate objective is being fulfilled [10].

3. Limited powers of Data Protection Authority in comparison with the Central Government

In comparison with the last version of the Personal Data Protection Bill, 2018 prepared by the Committee of Experts led by Justice Srikrishna, we witness an abrogation of powers of the Data Protection Authority (Authority), to be created, in this Bill. The powers and functions that were originally intended to be performed by the Authority have now been allocated to the Central Government. For example:

In the 2018 Bill, the Authority had the power to notify further categories of sensitive personal data. Under the present Bill, the Central Government in consultation with the sectoral regulators has been conferred the power to do so.
Under the 2018 Bill, the Authority had the sole power to determine and notify significant data fiduciaries, however, under the present Bill, the Central Government has in consultation with the Authority been given the power to notify social media intermediaries as significant data fiduciaries.

In order to govern data protection effectively, there is a need for a responsive market regulator with a strong mandate and resources. The political nature of the personal data also requires that the governance of data, particularly the rule-making and adjudicatory functions performed by the Authority are independent of the Executive.

4. No clarity on data sandbox

The Bill contemplates a sandbox for “ innovation in artificial intelligence, machine-learning or any other emerging technology in public interest.” A Data Sandbox is a non-operational environment where the analyst can model and manipulate data inside the data management system. Data sandboxes have been envisioned as a secure area where only a copy of the company’s or participant companies’ data is located [11]. In essence, it refers to the scalable and creation platform which can be used to explore an enterprise’s information sets. On the other hand, regulatory sandboxes are controlled environments where firms can introduce innovations to a limited customer base within a relaxed regulatory framework, after which they may be allowed entry into the larger market after meeting certain conditions. This purportedly encourages innovation through the lowering of entry barriers by protecting newer entrants from unnecessary and burdensome regulation. Regulatory sandboxes can be interpreted as a form of responsive regulation by governments that seek to encourage innovation – they allow selected companies to experiment with solutions within an environment that is relatively free of most of the cumbersome regulations that they would ordinarily be subject to, while still subject to some appropriate safeguards and regulatory requirements. Sandboxes are regulatory tools which may be used to permit companies to innovate in the absence of heavy regulatory burdens. However, these ordinarily refer to burdens related to high barriers to entry (such as capital requirements for financial and banking companies), or regulatory costs. In this Bill, however, the relaxing of data protection provisions for data fiduciaries would lead to restrictions of the privacy of individuals. Limitations to a fundamental rights on grounds of ‘fostering innovation’ is not a constitutional tenable position, and contradict the primary objectives of a data protection law.

5. The primacy of ‘harm’ in the Bill ought to be reconsidered

While a harms based approach is necessary for data protection frameworks, such approaches should be restricted to the positive obligations, penal provisions and responsive regulation of the Authority. The Bill does not provide any guidance on either the interpretation of the term ‘harm,’ [12] or on the various activities covered within the definition of the term. Terms such as ‘loss of reputation or humiliation’ ‘any discriminatory treatment’ are a subjective standard and are open to varied interpretations. This ambiguity in the definition will make it difficult for the data principal to demonstrate harm and for the DPA to take necessary action as several provisions are based upon harm being caused or likely to be caused.Some of the significant provisions where ‘harm’ is a precondition for the provision to come into effect are —

Clause 25: Data Fiduciary is required to notify the Authority about the breach of personal data processed by the data fiduciary, if such breach is likely to cause harm to any data principal. The Authority after taking into account the severity of the harm that may be caused to the data principal will determine whether the data principal should be notified about the breach.
Clause 32 (2): A data principal can file a complaint with the data fiduciary for a contravention of any of the provisions of the Act, which has caused or is likely to cause ‘harm’ to the data principal.
Clause 64 (1): A data principal who has suffered harm as a result of any violation of the provision of the Act by a data fiduciary, has the right to seek compensation from the data fiduciary.

Clause 16 (5): The guardian data fiduciary is barred from profiling, tracking or undertaking targeted advertising directed at children and undertaking any other processing of personal data that can cause significant harm to the child.

6. Non personal data should be outside the scope of this Bill

Clause 91 (1) states that the Act does not prevent the Central Government from framing a policy for the digital economy, in so far as such policy does not govern personal data. The Central Government can, in consultation with the Authority, direct any data fiduciary to provide any anonymised personal data or other non-personal data to enable better targeting of delivery of services or formulation of evidence based policies in any manner as may be prescribed.It is concerning that the data protection bill has specifically carved out an exception for the Central Government to frame policies for the digital economy and seems to indicate that the government plans to freely use any and all anonymized and/or non-personal data that rests with any data fiduciary that falls under the ambit of the bill to support the digital economy including for its growth, security, integrity, and prevention of misuse. It is unclear how the government, in practice, will be able to compel organizations to share this data. Further, there is a lack of clarity on the contours of the definition of non-personal data and the Bill does not define the term. It is also unclear whether the Central Government can compel the data fiduciary to transfer/share all forms of non-personal data and the rights and obligations of the data fiduciaries and data principals over such forms of data. Anonymised data refers to data which has ‘ irreversibly’ been converted into a form in which the data principal cannot be identified. However, as several instances have shown ‘ irreversible’ anonymisation is not possible. In the United States, the home addresses of taxi drivers were uncovered and in Australia individual health records were mined from anonymised medical bills [13]. In September 2019, the Ministry of Electronics and Information Technology, constituted an expert committee under the chairmanship of Kris Gopalkrishnan to study various issues relating to non-personal data and to deliberate over a data governance framework for the regulation of such data.The provision should be deleted and the scope of the bill should be limited to protection of personal data and to provide a framework for the protection of individual privacy. Until the report of the expert committee is published, the Central Government should not frame any law/regulation on the access and monetisation of non-personal/ anonymised data nor can they create a blanket provision allowing them to request such data from any data fiduciary that falls within the ambit of the bill. If the government wishes to use data resting with a data fiduciary; it must do so on a case to case basis and under formal and legal agreements with each data fiduciary.

7. Steps towards greater decentralisation of power

We propose the following steps towards greater decentralisation of powers and devolved jurisdiction —

Creation of State Data Protection Authorities: A single centralised body may not be the appropriate form of such a regulator. We propose that on the lines of central and state commissions under the Right to Information Act, 2005, state data protection authorities are set up which are in a position to respond to local complaints and exercise jurisdiction over entities within their territorial jurisdictions.
More involvement of industry bodies and civil society actors: In order to lessen the burden on the data protection authorities it is necessary that there is active engagement with industry bodies, sectoral regulators and civil society bodies engaged in privacy research. Currently, the Bill provides for involvement of industry or trade association, association representing the interests of data principals, sectoral regulator or statutory Authority, or an departments or ministries of the Central or State Government in the formulation of codes of practice. However, it would be useful to also have a more active participation of industry associations and civil society bodies in activities such as promoting awareness among data fiduciaries of their obligations under this Act, promoting measures and undertaking research for innovation in the field of protection of personal data.

8. The Authority must be empowered to exercise responsive regulation

In a country like India, the challenge is to move rapidly from a state of little or no data protection law, and consequently an abysmal state of data privacy practices to a strong data protection regulation and a powerful regulator capable of enabling a state of robust data privacy practices. This requires a system of supportive mechanisms to the stakeholders in the data ecosystem, as well as systemic measures which enable the proactive detection of breaches. Further, keeping in mind the limited regulatory capacity in India, there is a need for the Authority to make use of different kinds of inexpensive and innovative strategies.We recommend the following additional powers for the Authority to be clearly spelt out in the Bill —

Informal Guidance: It would be useful for the Authority to set up a mechanism on the lines of the Security and Exchange Board of India (SEBI)’s Informal Guidance Scheme, which enables regulated entities to approach the Authority for non-binding advice on the position of law. Given that this is the first omnibus data protection law in India, and there is very little jurisprudence on the subject from India, it would be extremely useful for regulated entities to get guidance from the regulator.
Power to name and shame: When a DPA makes public the names of organisations that have seriously contravened data protection legislation, this is a practice known as “naming and shaming.” The UK ICO and other DPAs recognise the power of publicity, as evidenced by their willingness to co-operate with the media. The ICO does not simply post monetary penalty notices (MPNs or fines) on its websites for journalists to find, but frequently issues press releases, briefs journalists and uses social media. The ICO’s publicity statement on communicating enforcement activities states that the “ICO aims to get media coverage for enforcement activities.”
Undertakings: The UK ICO has also leveraged the threats of fines into an alternative enforcement mechanism seeking contractual undertakings from data controllers to take certain remedial steps. Undertakings have significant advantages for the regulator. Since an undertaking is a more “co-operative”solution, it is less likely that a data controller will change it. An undertaking is simpler and easier to put in place. Furthermore, the Authority can put an undertaking in place quickly as opposed to legal proceedings which are longer.

9. No clear roadmap for the implementation of the Bill

The 2018 Bill had specified a roadmap for the different provisions of the Bill to come into effect from the date of the Act being notified [14]. It specifically stated the time period within which the Authority had to be established and the subsequent rules and regulations notified.The present Bill does not specify any such blueprint; it does not provide any details on either when the Bill will be notified or the time period within within which the Authority shall be established and specific rules and regulations notified. Considering that 25 provisions have been deferred to rules that have to be framed by the Central Government and a further 19 provisions have been deferred to the regulations to be notified by the Authority the absence and/or delayed notification of such rules and regulations will impact the effective functioning of the Bill.The absence of any sunrise or sunset provision may disincentivise political or industrial will to support or enforce the provisions of the Bill. An example of such a lack of political will was the establishment of the Cyber Appellate Tribunal. The tribunal was established in 2006 to redress cyber fraud. However, it was virtually a defunct body from 2011 onwards when the last chairperson retired. It was eventually merged with the Telecom Dispute Settlement and Appellate Tribunal in 2017.We recommend that Bill clearly lays out a time period for the implementation of the different provisions of the Bill, especially a time frame for the establishment of the Authority. This is important to give full and effective effect to the right of privacy of the
individual. It is also important to ensure that individuals have an effective mechanism to enforce the right and seek recourse in case of any breach of obligations by the data fiduciaries.For offences, we suggest a system of mail boxing where provisions and punishments are enforced in a staggered manner, for a period till the fiduciaries are aligned with the provisions of the Act. The Authority must ensure that data principals and fiduciaries have sufficient awareness of the provisions of this Bill before bringing the provisions for punishment are brought into force. This will allow the data fiduciaries to align their practices with the provisions of this new legislation and the Authority will also have time to define and determine certain provisions that the Bill has left the Authority to define. Additionally enforcing penalties for offences initially must be in a staggered process, combined with provisions such as warnings, in order to allow first time and mistaken offenders from paying a high price. This will relieve the fear of smaller companies and startups who might fear processing data for the fear of paying penalties for offences.

10. Lack of interoperability

In its current form, a number of the provisions in the Bill will make it difficult for India’s framework to be interoperable with other frameworks globally and in the region. For example, differences between the draft Bill and the GDPR can be found in the grounds for processing, data localization frameworks, the framework for cross border transfers, definitions of sensitive personal data, inclusion of the undefined category of ‘critical data’, and the roles of the authority and the central government.

11. Legal Uncertainty

In its current structure, there are a number of provisions in the Bill that, when implemented, run the risk of creating an environment of legal uncertainty. These include: lack of definition of critical data, lack of clarity in the interpretation of the terms ‘harm’ and ‘significant harm’, ability of the government to define further categories of sensitive personal data, inclusion of requirements for ‘social media intermediaries’, inclusion of ‘non-personal data’, framing of the requirements for data transfers, bar on processing of certain forms of biometric data as defined by the Central Government, the functioning between a consent manager and another data fiduciary, the inclusion of an AI sandbox and the definition of state. To ensure the greatest amount of protection of individual privacy rights and the protection of personal data while also enabling innovation, it is important that any data protection framework is structured and drafted in a way to provide as much legal certainty as possible.

Endnotes

1. (2017) 10 SCC 641 (“Puttaswamy I”).

2. Clause 42(1) of the 2018 Bill states that “Processing of personal data in the interests of the security of the State shall not be permitted unless it is authorised pursuant to a law, and is in accordance with the procedure established by such law, made by Parliament and is necessary for, and proportionate to such interests being achieved.”

3. (2019) 1 SCC 1 (“Puttaswamy II”)

4. Puttaswamy I, supra, para 180.

5. (1978) 1 SCC 248.

6. Ibid para 48.

7. Puttaswamy I supra para 180.

8. State of W.B. v. Anwar Ali Sarkar, 1952 SCR 284; Satwant Singh Sawhney v A.P.O AIR 1967 SC1836.

9. (2016)7 SCC 353.

10. Dvara Research “Initial Comments of Dvara Research dated 16 January 2020 on the Personal Data Protection Bill, 2019 introduced in Lok Sabha on 11 December 2019”, January 2020, https://www.dvara.com/blog/2020/01/17/our-initial-comments-on-the-personal-data-protection-bill-2019/ (“Dvara Research”).

11. “A Data Sandbox for Your Company”, Terrific Data, last accessed on January 31, 2019, http://terrificdata.com/2016/12/02/3221/.

12. Clause 3(20) — “harm” includes (i) bodily or mental injury; (ii) loss, distortion or theft of identity; (ii) financial loss or loss of property; (iv) loss of reputation or humiliation; (v) loss of employment; (vi) any discriminatory treatment; (vii) any subjection to blackmail or extortion; (viii) any denial or withdrawal of service,benefit or good resulting from an evaluative decision about the data principal; (ix) any restriction placed or suffered directly or indirectly on speech, movement or any other action arising out of a fear of being observed or surveilled; or (x) any observation or surveillance that is not reasonably expected by the data principal.

13. Alex Hern “Anonymised data can never be totally anonymous, says study”, July 23, 2019 https://www.theguardian.com/technology/2019/jul/23/anonymised-data-never-be-anonymous-enough-study-finds.

14. Clause 97 of the 2018 Bill states“(1) For the purposes of this Chapter, the term ‘notified date’ refers to the date notified by the Central Government under sub-section (3) of section 1. (2)The notified date shall be any date within twelve months from the date of enactment of this Act. (3)The following provisions shall come into force on the notified date-(a) Chapter X; (b) Section 107; and (c) Section 108. (4)The Central Government shall, no later than three months from the notified date establish the Authority. (5)The Authority shall, no later than twelve months from the notified date notify the grounds of processing of personal data in respect of the activities listed in sub-section (2) of section 17. (6)The Authority shall no, later than twelve months from the date notified date issue codes of practice on the following matters-(a) notice under section 8; (b) data quality under section 9; (c) storage limitation under section 10; (d) processing of personal data under Chapter III; (e) processing of sensitive personal data under Chapter IV; (f ) security safeguards under section 31; (g) research purposes under section 45; (h) exercise of data principal rights under Chapter VI; (i) methods of de-identification and anonymisation; (j) transparency and accountability measures under Chapter VII. (7)Section 40 shall come into force on such date as is notified by the Central Government for the purpose of that section.(8)The remaining provision of the Act shall come into force eighteen months from the notified date.”

For more details visit https://cis-india.org/internet-governance/blog/comments-to-the-personal-data-protection-bill-2019

Call for Researchers: Welfare, Gender, and Surveillance

ambika — 2020-02-13T15:05:37Z

We are inviting applications for two researchers. Each researcher is expected to write a narrative essay that interrogates the modes of surveillance that people of LGBTHIAQ+ and gender non-conforming identities and sexual orientations are put under as they seek sexual and reproductive health (SRH) services in India. The researchers are expected to undertake field research in the location they are based in, and reflect on lived experiences gathered through field research as well as their own experiences of doing field research. Please read the sections below for more details about the work involved, the timeline for the same, and the application process for this call.

Call for Researchers: Download (PDF)

Description of the Work

Each researcher is expected to author a narrative essay that presents and reflects on lived experiences of people of LGBTHIAQ+ and gender non-conforming identities and sexual orientations as they seek sexual and reproductive health (SRH) services in India. We expect the essay to contribute to a larger body of knowledge around the increasing focus on data-driven initiatives for public health provision in the country and elsewhere. Accordingly, the researcher may respond to any one or more than one of the following questions, within the context of the geographical focus as specified by the researcher:

What are the modes of surveillance, especially in terms of generation and exploitation of digital data, experienced by people of marginalised gender identities and sexual orientations in India, as they avail of sexual and reproductive healthcare?
How are the lived experiences of underserved populations, such as people of marginalised gender identities and sexual orientations, shaped by gendered surveillance while accessing sexual and reproductive services?
What are the modes of governance and gender ideologies that have mediated the increasing datafication of such provision?

We expect the researchers to draw on a) the Indian Supreme Court’s framing of privacy in India, as a fundamental right, and its implications; and b) apply and/or build on feminist conceptualisations of privacy. Further, we expect the researchers to respond to the uncertain landscape of legal rights accessible to people of LGBTHIAQ+ and gender non-conforming identities and sexual orientations, especially in the current context shaped by The Transgender Persons (Protection of Rights) Act, 2019.

The researchers will undertake field research in locations of their choice, conduct interviews and discussions with people of LGBTHIAQ+ and gender non-conforming identities and sexual orientations seeking such services, and conduct formal and informal interviews with officials and personnel associated with public and private sector agencies involved in the provision of SRH services.

Eligibility and Application Process

We specifically encourage people of LGBTHIAQ+ and gender non-conforming identities and sexual orientations to submit their applications for this call for researchers.

We are seeking applications from individuals who:

Are based in the place where field study is to be undertaken, for the duration of the study;
Are fluent in the main regional language(s) spoken in the city where the study will be conducted, and in English (especially written);
Preferably have a postgraduate degree (current students should also apply) in social or technical sciences, journalism, or legal studies (undergraduate degree-holders with research or work experience should also apply); and
Have previous research and writing experiences on issues at the intersection of sexual and reproductive health, gender justice and women’s rights, and health informatics or digital public health.

Please send the following documents (in text or PDF formats) to raw@cis-india.org by Friday, January 24 to apply for the researcher positions:

Brief CV with relevant academic and professional information;
Two samples of academic/professional (published/unpublished) writing by the applicant; and
A brief research proposal (around 500 words) that should specify the scope (geographical and conceptual), research questions, and motivation of the essay to be authored by the applicant.

All applicants will be informed of the selection decisions by Friday, January 31.

Timeline of the Work

February 3-7 CIS research team will have a call with each researcher to plan out the work to be undertaken by them

February - March Researchers are to undertake field research, as proposed by the researchers and discussed with the CIS research team

March 27 Researchers are to submit a full draft essay (around 3,000 words)

March 30 - April 3 CIS research team will have call with each researcher to discuss the shared draft essays and make plans towards their finalisation

May 15 Researchers are to submit the final essay (around 5,000 words, without footnotes and references)

As part of this project, CIS will organise two discussion events in Bengaluru and New Delhi during April-June (tentatively). Event dates are to be decided in conversation with the researchers, and they will be invited to present their works in the same.

Remuneration

Each researcher will be paid a remuneration of Rs. 1,00,000 (inclusive of taxes) over two equal installments: first on signing of the agreement in February 2020, and second on submission of the final essay in May 2020.

We will also reimburse local travel expenses of each researcher upto Rs. 10,000, and translations and transcriptions expense (if any) incurred by each researcher upto Rs. 10,000. These reimbursements will be made on the basis of expense invoices shared by the researcher.

Description of the Project

Previous research conducted by CIS on the subject of sexual and reproductive health (SRH) services in India observes that there is a complex web of surveillance, or ‘dataveillance’, around each patient as they avail of SRH services from the state. In this current project, we are aiming to map the ecosystem of surveillance around SRH services as their provision becomes increasingly ‘data-driven’, and explore its implications for patients and beneficiaries.

Through this project, we are interested in documenting the roles played by both the public and the private sector actors in this ecosystem of health surveillance. We understand the role of private sector actors as central to state provision of sexual and reproductive health services, especially through the institutionalisation of data-driven health insurance models, as well as through extensive privatisation of public health services. By studying semi-private, private, and public medical establishments including hospitals, primary/community health centres and clinics, we aim to develop a comparative analysis of surveillance ecosystems across the three establishment types.

This project is led by Ambika Tandon, Aayush Rathi, and Sumandro Chattapadhyay at the Centre for Internet and Society, and is supported by a grant from Privacy International.

Indicative Reading List

We are sharing below a short and indicative list of readings that may be useful for potential applicants.

Aayush Rathi, Is India's Digital Health System Foolproof? (2019)

Aayush Rathi and Ambika Tandon, Data Infrastructures and Inequities: Why Does Reproductive Health Surveillance in India Need Our Urgent Attention? (2019)

Ambika Tandon, Feminist Methodology in Technology Research: A Literature Review (2018)

Ambika Tandon, Big Data and Reproductive Health in India: A Case Study of the Mother and Child Tracking System (2019)

Anja Kovacs, Reading Surveillance through a Gendered Lens: Some Theory (2017)

Lindsay Weinberg, Rethinking Privacy: A Feminist Approach to Privacy Rights after Snowden (2017)

Nicole Shephard, Big Data and Sexual Surveillance (2016)

Sadaf Khan, Data Bleeding Everywhere: A Story of Period Trackers (2019)

For more details visit https://cis-india.org/jobs/researchers-welfare-gender-surveillance-call

IETF106

Admin — 2019-12-15T06:14:02Z

Gurshabad Grover participated at IETF106, which was held in Singapore 16-22 November, 2019.

In the meeting of the Human Rights Protocol Considerations (hrpc) research group, I presented an update to draft-irtf-hrpc-guidelines-03 (Guidelines for Human Rights Protocol and Architecture Considerations), which is an Internet Draft adopted by the hrpc rg that he is co-editing with Niels ten Oever. More info here.

Among other working/research group meetings, I participated theTransport Layer Security (tls) and the Privacy Enhancements and Assessments research group (pearg) sessions. I also participated inseveral side meetings, including the Public Interest Technology Group(pitg) meeting.

Agenda for the IETF and the different WGs/RG can be found on the IETF website.

For more details visit https://cis-india.org/internet-governance/news/ietf106

Power over privacy: New Personal Data Protection Bill fails to really protect the citizen’s right to privacy

Nikhil Pahwa — 2019-12-15T05:57:31Z

Nikhil Pahwa throws light on the new personal data protection bill.

The article by Nikhil Pahwa was published in the Times of India on December 12, 2019. CIS report was mentioned.

Earlier this year, in April, a data breach in the Election Commission of Philippines led to the leakage of personal information of over 55 million eligible voters on a searchable website: including names, addresses and date of birth. This was not the first data breach from the Election Commission. After the first, which took place in March 2016, where 340 GB of voter data was published online by a group of hackers called LulzSec Pilipinas, the National Privacy Commission of Philippines found that the Election Commission had violated the Data Privacy Act of 2012, and recommended criminal prosecution of its chairman, finding him liable when the agency failed to dispense its duty as a “personal information controller”.

It’s 2019, and that recommendation has still not been acted upon, because the National Privacy Commission of Philippines only has recommendatory powers for criminal prosecution. Meanwhile, data breaches continue at the Election Commission of Philippines.

Between 2017 and 2018, Aadhaar related personally identifiable data of several Indian citizens, including names, addresses, bank account numbers, in some cases pregnancy information and even religion and caste information of individuals, was published online by Indian government departments. The Centre for Internet and Society, in a report, estimated that personally identifiable data for 130-135 million Indian citizens had been leaked, thus putting them at risk. 210 government websites had made Aadhaar related data public, UIDAI confirmed in response to an RTI in 2017.

No one was held liable. There was no data protection law, no data protection authority, no criminal prosecution was recommended. Around that time, the Indian government was instead arguing in the Supreme Court that privacy isn’t a fundamental right under the Indian Constitution.

What we can learn from these two instances is that for the enforcement of a citizen’s right to privacy, and ensuring that no one takes the protection of data lightly, there needs to be a strong privacy law that holds even the government responsible, and above all, a strong data protection authority that is independent and has powers to penalise even government officials. On some of these counts, the Personal Data Protection Bill, 2019, disappoints.

First, members of the Data Protection Authority will no longer be appointed by independent entities from diverse backgrounds: where they were previously going to be appointed by a committee comprising the Chief Justice of India or a Supreme Court judge, the Cabinet secretary, and an independent expert, the power to appoint members to DPA now rests solely with government officials, including the appointment of adjudicating officers. In addition, the central government, in the interest of “national security, sovereignty, international relations and public order, can issue directions to DPA, which DPA will be bound by. Powers of DPA have also been reduced: while in the previous version of the bill, DPA had the sole power to categorise data as sensitive personal data, in the current version, the power rests with the central government, albeit in consultation with DPA. The central government will also notify any social media company as a significant data fiduciary, and not DPA. Only the central government can determine what critical personal data is, and not DPA.

This dependence on the government for appointments, functions and definitions, will invariably impact the independence of DPA, and even though the 2019 version of the bill gives it the authority to fine the state a maximum of Rs 5-15 crore, depending on the offence, i’d be surprised if this ever happens.

The bill does create significant exceptions for the state to acquire and process data, and an opportunity to create a base for surveillance reform in the country has been lost. The previous version of the bill had brought some sense of safety against mass surveillance, when it included the condition that processing of data by the government must be “necessary and proportionate”, drawing from Supreme Court’s historic right to privacy judgment. This is particularly important given that the bill also gives power to the government to exempt any agency from the provisions of the bill for processing of personal data, which includes acquiring data from any public or private entity.

Effectively, this means that government agencies may be exempt from any scrutiny by DPA, and can even collect data from third parties (for example, fin-tech companies, health-tech startups) without the user even knowing. Forget recommending criminal prosecution for mass surveillance, India’s DPA won’t even be able to fine a government agency for such a violation of the fundamental right to privacy. The government also has vast exceptions for data processing: “for the performance of any function of the state authorised by law”.

This aside, one of the more curious clauses in the bill is around non-personal data. The government, a few months ago, constituted a committee led by Infosys co-founder Kris Gopalakrishnan to look into the governance of non-personal data. Non-personal data, as the term suggests, is any data that is not related to an individual. In the bill, the government has given itself the right to acquire this data, which is essentially a company’s intellectual property, to “promote framing of policies for digital economy”. Why non-personal data finds a mention in a Personal Data Protection Bill is beyond comprehension, and this move will not inspire much confidence in businesses operating in India, when the state claims eminent domain over intellectual property.

It’s unfortunate minister Ravi Shankar Prasad is sending the bill to a select committee, given the fact that such significant changes to the bill should have led to another public consultation.

For more details visit https://cis-india.org/internet-governance/news/the-times-of-india-december-12-2019-power-over-privacy

Outrage As Privileged IITians Use Tech To Spy On Sweepers

Rachna Khaira — 2019-12-15T05:33:21Z

Some members of the housekeeping staff at IIT Ropar were put under round the clock surveillance during working hours for many days in February this year without their consent. IIT Ropar Director Prof S K Das has ordered a probe into the incident.

The article by Rachna Khaira was published in Huffington Post on December 31, 2019. Aayush Rathi was quoted.

The Indian Institute of Technology (IIT), Ropar is conducting a probe into the reported tagging and round the clock electronic surveillance of some housekeeping staff members as part of an experiment run by the Technology Business Incubation Foundation (TBIF) located at the IIT campus in February this year.

HuffPost India has learnt that the TBIF, a tech incubator run within IIT Ropar, signed off on the “Sweepy” project in which housekeeping staff were given wristbands and brooms secretly embedded with tracking chips, without seeking the consent of the janitorial staff, or informing IIT Ropar management.

While the housekeeping staff were told the wristbands would record their pulse and heart beat, and that they should wear it while cleaning the campus, the tracking chips were used to track to assess if they were sweeping out hard-to-reach corners of the institute.

Prof. Sarit Kumar Das, Director IIT Ropar told HuffPost India that a three member committee comprising of Prof. Bijoy H Barua, Prof. Javed Agrewala and Prof. Deepak Kashyap has been set up to look into the matter.

“We at the IIT Ropar respect privacy and condemn any such violation made by any of our student or staff member,” said Prof. Das. “Before conducting any experiment on human beings, an approval has to be sought from the human ethics team constituted in our institution and they present a case to me after seeking a written consent from the people who would undergo the experiment. Only, after getting my approval, such an experiment can be conducted at the campus.”

Sweeping surveillance

J K Sharma, the Chief Operating Officer of TBIF, told HuffPost India that his tech incubator deliberately misled the housekeeping staff about the true purpose of the wristband as they felt the housekeeping staff wouldn’t agree to wear such a device.

While elaborating more on the ‘Sweepy’ project, Sharma said that the project was based on an idea that came to the hostellers who were upset over the housekeeping staff for not cleaning their rooms.

“The sweepers were not working properly and despite reporting the matter several times to the authorities, they were not taking any cognisance. Perturbed, the students developed this programme in which the location of the sweeper can be recorded and monitored in a control room by a gadget tied to the sweeper’s wrist,” said Sharma.

He further added that a beacon records the activity of the sensor pasted to the broom or mop held by the sweeper and can monitor the area and the time in which it was used. The report was produced digitally on the screen.

Was a consent sought from the sweepers before tagging them?

“The testing was done in a secret manner as the housekeeping staff may not have given their consent for the trial. We tried it on three sweepers and while two of them were found working dedicatedly, one was found to have missed cleaning from few areas assigned to him,” said Sharma.

The findings were shared with the housekeeping supervisor who later directed his staff to do their duty more diligently.

The team working on the project however told HuffPost India that they secured the privacy of the housekeeping staff by removing the microphone from the gadgets tied to their wrists.

This technology does not have video feature and only monitors location of a moving object and is quite cheap as compared to the radio-frequency identification (RFID) technology that uses electromagnetic fields to automatically identify and track tags attached to objects.

The testing was done in a secret manner as the housekeeping staff may not have given their consent for the trial. We tried it on three sweepers and while two of them were found working dedicatedly, one was found to have missed cleaning from few areas assigned to himJ K Sharma, Chief Executive Officer, Technology Business Incubation Foundation, IIT Ropar

Calling this an increasingly commonplace trend of covert spying on domestic workers without their knowledge, Ayush Rathi, Programme Officer, Centre for Internet and Society, said that the housekeeping staff was made to wear the gadget under a false pretense is telling.

“This is a classic example of how the access to privacy is stratified along the axes of class, caste and gender. And ties in closely with a key purpose of surveillance — that of exerting control over people’s bodies to conform to the surveiller’s ideas of right and wrong,” said Rathi.

He further added that in many ways, this story captures the zeitgeist of the 21st century. The is the essence of so much of what qualifies as innovation today is that they seek to find technological solutions to problems that are structural in nature.

“So, in this instance it is very evident that the objective sought to be achieved was not to merely ‘fix’ the problem of the housekeeping staff performing its duties well, but to solely hold them guilty for failing to do so,” said Rathi.

An alternate, albeit more tedious, approach would have been to speak with the workers and iron out the struggles they were facing at the workplace that were preventing them from performing their job well. Any solution could only have been prepared thereafter — he added.

As per Prof. Das, a major problem with the engineering students is that unlike medical students, 90 percent of their experiments are based on machines and not human beings.

“There is too much deficiency of the understanding of human psychology amongst engineering students. To curb this, we at the IIT have started a mandatory course on human ethics which is being taught by some of the renowned human psychology experts. Still sometimes, the violations gets reported,” said Prof. Das.

For more details visit https://cis-india.org/internet-governance/news/huffignton-post-december-13-2019-rachna-khaira-outrage-as-privileged-iit-ians-use-tech-to-spy-on-sweepers