GDC online submission – final draft

as of Oct 14, 2022

Download as pdf the submission by the fellows of the 2022 European Summer School on Internet Governance (EuroSSIG) to the Global Digital Compact.

Overview

The European Summer School on Internet Governance (EuroSSIG), held annually, often includes an exercise for the fellows that tackles one of the important Internet Governance issues of the day. In 2022, the challenge posed to the fellows was to come up with a submission to the United Nations Office of the Secretary General’s Envoy on Technology invitation for comments on the United Nations Secretary General’s Global Digital Compact. Over the course of a week, the fellows, in their own capacities, participated in a multistakeholder exercise to develop a response to the call. The fellows came from a set of diverse stakeholder groups and had a wide range of geographical origins and residences.

The fellows were given the choice of which issues to address. They reached consensus on addressing 6 of the 7 areas, excluding the Area 5 “Accountability for discrimination and misleading content.”

Description of entity/organization

The European Summer School for Internet Governance (EuroSSIG) aims at providing a multi stakeholder learning environment for graduate students and young academics as well as junior professionals from private sector, government and civil society. All fellows come with some learning and often experience in some are related to the digital economy or Internet Governance. Since its inception in 2007 363 fellows from 97 countries aged between 20 to 70 years attended the European Summer School on Internet Governance. Each year, the fellows tackle one of the pressing issues of Internet Governance in an extended process. The group participating in 2022 that produced this submission came from 15 countries and came from all stakeholder groups.

Process followed to collect, consult, and prepare your input

Before the start of the program, the fellows, with diverse stakeholder backgrounds, were asked to read the “Global Digital Compact” and review “Our Common Agenda”, as well as other selected other relevant material. The fellows were asked to each come prepared with one core principle and a corresponding key commitment for at least one of the areas defined in the proposed compact to contribute to the initial session. Sessions were held each day for five days to discuss the issues and come to consensus. Some sessions were held in plenary with all fellows attending, while some were held within thematic issue groupings. Some bilateral sessions were held among the diverse stakeholder groups. Faculty at the school served as subject matter experts to advise, but were cautioned against directing the process. The entire process was run under predefined multistakeholder modalities.

Contributing text came from rapporteurs chosen within each of the thematic areas and was organized by two faculty members serving as the secretariat. Successive drafts were discussed in plenary sessions. Only consensus text moved forward. Consensus was defined as text to which no fellow objected after discussion and revision. The final submission was prepared by the secretariat and was reviewed and edited by the rapporteurs from each of the thematic areas. No new contributions were added after the final session when consensus was reached.

It should be noted that all opinions discussed during the process of developing the contribution were made in the personal capacity of the participants and did not necessarily reflect the position of their educational institutions or their employers. Any work done by EuroSSIG faculty, including the secretariat, was in support of the project and does not indicate support by them or by organisations with which they are affiliated.

The following sections contain the consensus response of the fellows at the EuroSSIG on July 22.

1. Connect all people to the Internet, including all schools

  1. Core Principles

    1. All people must be able to have meaningful Internet access, with special attention to affordability, digital skills, digital security, and social indicators, including but not restricted to gender, race, ethnicity.
    1. In line with the recommendation of the Alliance for Affordable Internet (A4AI), adopted by the UN Broadband Commission,‘meaningful Internet access’ is defined as being able to use the Internet, open and unfiltered, on a daily basis, with an appropriate device, including enough data (when it is a mobile connection). Connections should be provided with technical minimum requirements such as speed and latency. It should offer a realistic but clear threshold for many low- and middle-income countries, and look at the users on this network, not just the network’s coverage.
    1. In order to make people’s access less dependent on individual subscriptions at all times, it is important that all schools in underdeveloped areas are provided with meaningful Internet access, either via fixed line or satellite if local infrastructure is not available. Wifi access can then be offered as a public service to local citizens. Internet capacity per school should meet the demand of students and other local citizens using its Wi-Fi access.

  2. Key Commitment/ Pledges/ Actions

    1. Because of disparities in Internet coverage, access and usage will differ between regions. States should pursue partnerships with intergovernmental organizations and other stakeholders in order to foster the creation of digital inclusion strategies at the local level.
    2. Governments should ensure a competitive market environment. This action includes the adoption of legislation that fosters the development of network providers, such as ISPs and IXPs for underserved communities. For the economic viability that keeps small providers, governmental stimuli are essential in order to complement the counterparts of large providers in the expansion of Internet access. Governments need to ensure that an independent national regulatory authority is installed with sufficient funding, overseeing the telecommunications market and securing fair competition and, as a result, fair pricing for Internet access.
    3. In line with the recommendation adopted by the UN Broadband Commission, we propose to use the “1 for 2” formula as a way to measure affordability of Internet access: affordable Internet is where 1GB of mobile broadband data is priced at 2% or less of average monthly income.
    4. Technical and scientific communities should assess the government’s core Internet usage indicators periodically, in order to identify improvements.
    5. To increase the digital skills of Internet users, governments should make digital skills and security activities compulsory in the primary school curriculum.
    6. Private sector should be encouraged to hold training sessions in digital skills and security, especially for the marginalized populations and those in the most remote areas of their countries.
    7. Civil society organizations should be encouraged and supported on the creation of permanent awareness raising campaigns addressing the relevance of meaningful access targeting groups that remain unconnected, such as, but not limited to, women and girls.

2. Avoid Internet fragmentation

  1. Core Principles

    1. The Internet, being a network of networks, must remain interoperable.
    2. It is crucial that we, as a global community, preserve the integrity of the Internet.
    3. The Internet must remain globally reachable through a unique identifier system.
    4. Internet openness is vital as it allows to achieve potential benefits for all.
    5. The efforts to prevent Internet fragmentation should take place simultaneously on three levels: on the technical layer, the regulatory layer, and socio-economical layer.

    The interoperable and globally reachable Internet, as we have known it since its origin, has enabled enormous network effects. The positive impact on socio-economic development, citizen participation, access to information and all kinds of digital services has completely changed our societies. Preserving a single, interoperable, open and globally reachable Internet is therefore in the global public interest.

  2. Key Commitment/ Pledges/ Actions

    Governments should:

    1. respect the public core of the Internet and pledge not to abuse the Internet’s infrastructure.
    2. ensure that their actions and policies do not negatively affect the core principles of the Internet.
    3. establish clear and transparent frameworks and due process when implementing laws and regulations on Internet utilization.
    4. exempt the technical community from Internet sanctions that infringe on the Internet infrastructure layer.
    5. ensure that their actions and regulations preserve the right to create, distribute, and access information resources in an interoperable network.

    Private Sector should:

    1. respect the public core of the Internet and pledge not to abuse the Internet’s infrastructure.
    2. implement due process, proportionality, and transparent practices for traffic management and content blocking in accordance with applicable legislations.
    3. ensure that in the provision of its services it does not create artificial barriers in Internet access and usage.

    Technical Community should:

    1. in its key role in preserving the unique identifier system, keep working collaboratively in maintaining and advancing the technology of the global Internet infrastructure.
    2. implement actions to enhance the resiliency and robustness of the network in order to prevent fragmentation attempts.

    Civil Society should:

    1. focus its actions on educating users and civil servants, as well as raising awareness about the threats that may be caused by Internet fragmentation in a constructive manner.
    2. take more efforts in monitoring and reporting on the fragmentation of the Internet.
    3. hold governments, the private sector, and the technical community accountable for their commitments to prevent the fragmentation of the Internet and for abuse of the Internet infrastructure.

    Academia:

    1. We encourage academia to contribute to the research and assessment efforts around the state of Internet fragmentation, as well as develop the system of indexes and measurements.
    2. We urge academic institutions to develop more educational programmes on Internet Governance and create specific courses of Internet openness.

    All stakeholder groups should

    engage in meaningful multistakeholder dialogue to prevent the fragmentation of the Internet.

3. Protect data

  1. Core Principles

    1. Privacy protection is a right of any citizen to control their own personal information and to decide about it in the digital space.
    2. Data subjects should have the right to understand how personal data is processed in an understandable way for all. Data subjects for which processing is explicitly agreed upon without any type of discrimination shall have be treated transparently.
    3. Prior to obtaining personal information, the potential processor of the information must provide the data subject with a complete understanding of its use in a manner understandable to all parties.
    4. Data processors and controllers should explain to data subjects, in accessible and simple language, why and how their personal data is processed, in observance of the principle of explainability.
    5. Personal data is personal and not a commodity by nature. Personal data can never be used by any actor by default. Where required, explicit consent must be given by the personal data subject prior to usage by third parties.
    6. To obtain personal information, the potential processor of personal data must be transparent towards the personal data subject about the entire data value chain (collection-storage-reuse-transfer) with no discrimination.
    7. Violation of core principles of data protection should be subject to penalties and sanctions.

  2. Key Commitment/ Pledges/ Actions

    Governments should:

    1. approve and enforce human-centric privacy and data protection frameworks to empower data subjects and protect individuals from abuses and misuses in the processing of their data.
    2. ensure that legal systems provide for accountability for abuses and misuses of personal data for any data subject under any circumstances.
    3. create and provide suitable and independent competences to Data Protection Agencies as those responsible for auditing and evaluating the enforcement of laws.
    4. ensure that curricula in primary and secondary education contain sufficient material on data usage and data privacy, cybersecurity and protection against cybercriminality, and awareness and understanding of misinformation and fake news.

    Private Sector should:

    1. ensure the protection of personal information.
    2. disclose their data management practice pertaining to personal data usage.
    3. ensure full transparency of their data value chain.
    4. ensure that no discriminatory practices occur towards any personal data subject, particularly vulnerable groups.
    5. along with the technical community, take responsibility for designing systems processing personal data, responsibly architect, and customize personal data protection policy to meet the needs of individual jurisdictions. Cost should be proportionate to the circumstances of the personal data processed.
    6. ensure that limitations to personal data processing does not limit access to necessary digital services.
    7. provide training in digital skills, cybersecurity, and data management to employees to ensure they are in compliance with laws and regulations.

    Technical Community should:

    1. take data protection principles into account when designing technical systems such as new technologies and online payment systems involving personal data, particularly in APPs.
    2. along with the private sector, design systems that process personal data in a responsible and secure way.
    3. along with the private sector, ensure that development of new technologies respects the core data protection principles.

    Civil Society:

    1. has the indiscriminate and unconditional right to know how personal data is treated through the entire data value chain (collection-storage-reuse-transfer).
    2. By default, data subjects should not be considered as having given the consent to provide their personal data by consenting to access to digital services. Civil society should have the right to access basic service functionality regardless of such consent.
    3. should create and disseminate information about data usage and data privacy, cybersecurity and measures against cyber criminality and awareness and understanding of misinformation and fake news, to encourage continuous and lifelong awareness of rights and empowerment of data subjects.

    Academia is encouraged:

    1. to organise/establish fora where all stakeholders could discuss issues, practices, and lessons without being driven by interests.
    2. to further research stakeholders’ relations, interests, and balance to enrich the community with results, which support the achievements and enhance core principles.
    3. to create curricula on digital literacy and skills at all levels and different formats (formal and informational). These would cover: a) awareness of managing and protecting personal data with the aim to prevent abuses, leaks, and breaches; b) user security and self awareness to avoid being a victim of cybercrime; c) awareness and understanding of misinformation and fake news.

4. Apply human rights online

  1. Core Principles

    1. It is imperative that the application of Human Rights to the digital space ensures that the rights we enjoy offline should also be enjoyed online. That means all stakeholders must commit to discussing and implementing measures and policies with regards to the following: allow everyone to have access to the Internet and information, address the gender digital divide, building confidence and trust in the Internet, eliminating all forms of discrimination, and also acknowledging privacy and freedom of expression need to be ensured in the online realm.

    Promoting equal participation through access to the Internet

    1. Promote universal and equal access to a multilingual Internet and ICT applications in order to encourage economic development, education and culture for all sectors, especially the most disadvantaged. Meaningful access must be perceived as means for full political participation and exercise of rights, especially for women and girls, LGBTQIA+, people with disabilities, communities with different ethnical backgrounds and other minorities.

    Addressing and combating gender-based violence

    1. All human rights violations and abuse facilitated through the use of the Internet and new technologies should be condemned and addressed, including all acts of sexual and gender-based violence committed by all parties. Online gender-based violence has a real and long lasting effect on the victims lives and, if not addressed, could result in a chilling effect on freedom of expression and the online existence of women, girls and gender-diverse groups.

    Right to privacy

    1. Ensure every citizen has the right to privacy and data protection in the digital space as means of empowering data subjects against the abuses and data misuses prompted by new technologies. Responsibly apply the use of technologies and provide checks and balances with regards to state authority activities in the national security realm.

    Freedom of expression

    1. Address the threats to the right to freedom of expression posed by the growing power of digital platforms, as well as censorship and filtering activities performed by States. Ensure the existence of local frameworks dedicated to protecting the fundamental right to freedom of expression in the digital age.

  2. Key Commitment/ Pledges/ Actions

    Promoting equal participation through access to the Internet

    States should:
    1. Develop medium-term action plans aimed at reducing the digital divide, with the special focus on women and girls, LGBTQIA+, people with disabilities, communities with different ethnical backgrounds and other minorities.
    2. Adopt mechanisms for following up and evaluating the implemented initiatives.
    3. Invest in capacity building and digital literacy related initiatives.
    4. Publish reports on the advancement of access to the Internet and ICT applications at the national level, ideally on an annual basis.

    Right to privacy

    States should:
    1. Establish and enforce data protection frameworks with the data subjects at the very core and that are responsible for (a) defining the legal basis and remedies and (b) creating Data Protection Agencies responsible for the implementation and oversight of the legislation;
    2. Ensure the appropriate means for oversight of governmental intelligence and law enforcement activities through judicial and administrative mechanisms as well as increased transparency about the deployment of such technologies and the legal basis. Additionally, establish entities to evaluate the implementation of the above mentioned and other technologies responsible for the collection of citizens personal data;
    3. Perform periodic Human Rights and Surveillance impact assessments on the deployment of surveillance and less pervasive technologies (OSINT) by law enforcement and intelligence agencies;
    4. Limit the purchase and export of surveillance technology, as well as the implementation of less pervasive technologies that can be applied for the same purposes;
    5. Ensure that surveillance and intelligence activities are not retroactively authorized nor deployed, and abide by the principles of legality, necessity and proportionality.
    Private Sector should:
    1. Commit to protecting users from being personally targeted by digital threats, such as spyware and other forms of least pervasive technologies.
    2. Avoid the development of facial recognition or other security related databases with biased datasets or without the performance of human rights and surveillance impact assessments in order to avoid reinforcing discrimination or racism.

    Freedom of expression

    States should:
    1. Responsibly consider the development of regulatory frameworks for online platforms in order to halt the abuses facilitated by the internal policies of such actors;
    2. Ensure the protection of citizens rights to access and interact with information and avoid deploying authoritarian measures such as Internet shutdowns;
    3. Impose transparency obligations to private actors such as social media platforms with regards to the implementation of content governance related measures. Additionally address the lack of access to platform data and promote a broader understanding of how these actors work, through the publication of periodic transparency reports
    4. Regulate the use of Artificial Intelligence applications in order to mitigate chilling effects on the right to freedom of expression caused by their deployment over content shared by Internet users online.
    Private Sector should:
    1. Commit to the publication of periodic transparency reports applied to their content governance related policies.
    2. Collaborate with other stakeholders prior to the creation of new policies that could represent a risk or a possible restriction to Freedom of expression online.

5. Accountability for discrimination and misleading content

The fellows did not develop a response on this issue.

6. Regulation of artificial intelligence

  1. Core Principles

    The fellows decided to submit a comment on AI and culture, while recognising that AI and its application represent a broader topic.

    AI AND CULTURE

    1. Artificial intelligence (AI) is becoming an increasingly more relevant development in the art world, as illustrated by DALL-E-2’s capabilities. Artificial Neural Networks (ANNs) in particular hold a potential to negatively disrupt the creative industries. ANNs use large collections of human-created artistic works, known as “training data”, often without permission or knowledge of the authors, to create an AI model capable of synthesizing novel works.
    2. Based on these developments and following Paragraph 99 of UNESCO’s Recommendation on the Ethics of Artificial Intelligence, which calls for new research at the intersection between AI and intellectual property, we suggest the following principles for the development and use of AI tools in art:
      1. Continued cultural flourishment. To preserve continued and diverse cultural creation for all future generations, the most important principle of AI in art is that it should be designed, developed, commercialized and used to assist human creators, and never to substitute or harm their livelihood.
      2. Maintenance of copyright in the hands of creators. Copyright must remain in the hands of human creators and access to AI tools must not involve abusive practices such as giving up intellectual property rights over creations.
      3. AI-created vs. AI-assisted. Copyright law must distinguish between works created by AI vs. by humans with AI assistance, and the second group must not be hindered.
      4. Copyright by AI training and exceptions. The copyright regimen for AI creations must respect the artists’ whose works were used to train the tools while having a robust and clear set of copyright exceptions, including but not limited to academic and educational purposes.
    3. Keeping in mind cultural and artistic diversity across the centuries, we call developers of AI art tools to, following the example of copyright law, adopt a principle of aesthetic neutrality — not imposing blanket bans or hindrances on specific kinds of artistic content that unfairly affect cultural practices, particularly those outside Western traditions. Content policies, while important for safety, must not impose any specific cultural standard.

    AI AND REGULATION

    1. Additionally, the AI development process calls for an inclusive and facilitative environment. It must involve a variety of key actors, most importantly the end users, who bear the brunt of the final outcome of such development and by the experience of whom the overall benefits or drawbacks of the AI is judged as a yardstick.

  2. Key Commitment/ Pledges/ Actions

    AI AND CULTURE

    1. Governments and international organizations should work towards the drafting and signing of a complement to the Berne Convention. This treaty should establish artists‘ rights and the regulations around the creation of artistic works by Artificial Intelligence algorithms, in particular ANNs.
      This treaty should cover (1) the rights of artists whose work is used to train AI algorithms, (2) the rights of those who might be materially harmed by AI-generated creations, and (3) the licensing and copyright status of AI-generated works.
      It should establish that:

      1. Artists have a right to know if their works were used to train an AI and, anticipating the creation of hateful content with AI and falsely attributed to authors, to “non-misattribution”;
      2. Works must disclose to what extent AI was used to create it, though exceptions may be acceptable to not needlessly restrict artistic freedom;
      3. AI-generated works derive copyright from the material used to train it and must enjoy a short period of protection. A license from right holders is necessary to train an AI with copyrighted material. The rights over the output belong, at least partly, to the authors whose works were used;
      4. Licensing of works for training cannot be done in Terms of Use contracts, or as a condition to access a platform, tool or service, and
      5. An AI-created work will belong to the public domain when the AI was trained with public domain works, unless meaningfully modified by a human.
    2. We highlight that, over the centuries, artists all over the world and of all cultures have portrayed nudity. Many of the most ancient, classical and culturally diverse works of art involve nude subjects. As such AI developers must not impose bans on it.
    3. We commend the efforts of AI developers to adopt content policies to mitigate risk of deep fakes and the creation of hateful content, and support the adoption of content policies for AI art tools. However, developers must not impose creative restrictions, and limitations must not go beyond the necessary and proportional to prevent harm.

    AI AND REGULATION

    1. Following the example of the IEEE Code of Ethics, the technical community should develop a code of ethics for all participants (developers, engineers, scientists, etc) in the AI ​​ecosystem. Recognizing the inherent unpredictability of AI, this code of ethics should follow best practices for an ethical and rights-based approach to AI principles and require best efforts to predict, identify, and mitigate discrimination and potential harm. It should be written in a language that can be referenced and understood by the entire technical community and easily implemented in practice, by using an interpretive framework like the one created by the IETF.

7. Digital commons as a global public good

  1. Core Principles

    Defining the digital commons and their status as global public goods:

    1. It is necessary to reach a common understanding of the status of certain digital commons, such as collaborative knowledge platforms and the digital public sphere, as global public goods.
    2. The digital commons fit the definition of a public good, a good with high intrinsic social value and societal demand. However, there is a disparity between this and other public goods.
    3. Crucially, digital commons are a public good that enables access to other public goods, such as education and healthcare.

    How inequities in the digital commons should be resolved:

    1. It is necessary to recognise that inequities in access and meaningful participation within the digital commons are shaped by social inequalities — relating to (but not limited to) gender, ethnicity, language, wealth, sexual orientation, immigration status, and disability status — and their relationship to the unequal distribution of technological affordances present within the digital commons (i.e. interfaces which only afford the use of specific languages, in practice excluding primary speakers of other languages from a platform).
    2. Our comment emphasizes the need to adopt a design justice lens for widening access to and participation in the digital commons, involving an approach which centers the perspectives of affected communities and intersectional analysis of technological affordances and disaffordances within the digital commons.

    How the digital commons should be governed:

    1. Bearing in mind market pressures to constrain the scope of collaborative knowledge platforms, it is fundamental to reach common understanding on the status of digital commons as a global public good.
    2. Understanding digital commons as global public goods, the governance of such must abide by the principles and the structure of multistakeholderism.
    3. The digital commons must also be protected against market pressures and governmental inference, guaranteeing that they can continue to exist and to be accessible in the future while also retaining their properties and status as global public goods.

  2. Key Commitment/ Pledges/ Actions

    Defining the digital commons and their status as global public goods:

    1. Academic and civil society stakeholder groups should be encouraged to develop a precise theoretical foundation of the digital commons as a global public good, through concrete investment from states and the private sector on research and dialogue.
    2. States and private sector groups should be encouraged to contribute resources towards the alignment of the governance of the digital commons with that of other public goods, with a specific focus on the preservation of the digital commons for future generations.

    How inequities in the digital commons should be resolved:

    1. Institutions in the digital commons (like collaborative knowledge platforms) should identify inequities within the digital commons by adopting both community consultations as well as intersectional and human rights focused analyses of such inequities. This should follow the model of prior studies of the gender gap on Wikipedia or unequal representations within the Creative Commons.
    2. Institutions in the digital commons should acknowledge and address the global dimension of their work by expanding translation and linguistic access, by creating clear timelines for localisation for different language groups.
    3. Institutions in the digital commons should provide simple and plain-language explanations of their terms and conditions, to ensure the terms of participating and co-ownership of the digital commons to users are legible.

    How the digital commons should be governed:

    1. The „Digital Commons Architecture“ (DCA) proposed by the United Nations must be taken into account when creating mechanisms for multistakeholder, global and inclusive stewardship of these commons. These DCA-based governance structures would foster collaboration and legal harmonization.
    2. Digital commons protection against market pressures and governmental interference can be enhanced by discussing the possibility of understanding the digital commons as global public goods in the development of treaties.
    3. The overarching duties and responsibilities present on those treaties could be modeled after previous documents such as the Outer Space Treaty. Developing a similar framework of principles and duties and expanding their adherence to various stakeholders would maximize the recognition of and therefore the necessary safeguards for digital commons.

Annex:
List of EuroSSIG 2022 Fellows signing onto the opinion

  • Arham Shabbir
  • Bastiaan Goslings
  • Berenice Fernandez Nieto
  • Bernardo Barbosa
  • Bruna Santos
  • Everton T. Rodrigues
  • Frank van Berkel
  • Giorgi Aladashvili
  • Gulalai Khan
  • Gustavo Paiva
  • Julia Lacerda
  • Louise Blandin
  • Mireia Paulo
  • Nadezhda Arteeva
  • Niharika Gujela
  • Paloma Rocillo
  • Philipp Schulte
  • Rocío de la Fuente
  • Ruchika Hirna
  • Shadi Alhakimi
  • Stephanie Teeuwen
  • Teo Kai Xiang
  • Uchenna Jerome Orji
  • Vladislav Ivanets