Developments in the internet governance environment January to March 2026

Wolfgang Kleinwächter Apr 15, 2026 29 min read

In the first quarter of 2026, the following four topics were at the forefront of the discussion on Internet governance:

The future of the IGF and the new WSIS architecture (WSIS+)
Artificial Intelligence
Cybersecurity
Global negotiations on the digital economy

I. The future of the IGF and the new WSIS architecture (WSIS+)

Following the decision by the 80th UN General Assembly on 20 December 2025 to grant the IGF permanent status within the UN and to continue the WSIS process until 2035 (WSIS+30), discussions on the implementation of the WSIS+20 Output Document (OD) and the future of the IGF began in January 2026. Two meetings of the IGF Multistakeholder Advisory Group (MAG) in February and March 2026 focused primarily on the modernisation of the WSIS architecture (WSIS+), the modalities of the IGF’s future work, the composition and tasks of the IGF’s governing bodies (MAG and Leadership Panel/LP), and IGF funding. The basis for these discussions included the reports of the IGF Leadership Panels[1] and the Informal Multistakeholder Sounding Boards (IMSB)[2], as well as several ‘non-papers’, including those from the MAG Working Group on Strategy[3] and the Swiss Government.

Integration of parallel UN processes

A key issue is the effective and resource-efficient integration of the now numerous parallel processes of global digital policy within a "WSIS+ architecture". The Swiss paper calls for close coordination between the WSIS action lines, the Global Digital Compact (GDC) and the UN Sustainable Development Goals (SDGs). The GDC review is due in 2027, the SDG review in 2030. Within this "WSIS+ architecture", there should be discussion and decision-making bodies that are interconnected and whose modalities are based on the São Paulo Multistakeholder Guidelines (SPMGs).

The IGF, as the "primary multi-stakeholder discussion platform", should focus on "issue identification, agenda framing and gathering input from the global community". The ITU’s WSIS Forum would primarily serve as the venue for discussions on the implementation of the WSIS Action Lines.
The increasing number of intergovernmental negotiations on digital issues should be coordinated by the UN Group on the Information Society (UNGIS). UNGIS, in turn, should ensure, through a newly formed Strategic Multistakeholder Advisory Group (SMAG), that the views of non-state stakeholders are adequately represented in the intergovernmental negotiations.
The outcomes of the various processes should then be discussed at the annual sessions of the UNCSTD (spring in Geneva) and the UN General Assembly (autumn in New York) and documented in the UN Secretary-General’s report, which is now to be submitted every two years.
Particular attention should be paid to close coordination between the IGF on the one hand and the new Global Dialogue on AI Governance and the UNCSTD Working Group on Data Governance (WGDG) on the other. Both the AI Dialogue and the WGDG are part of the GDC.
The WG-Strategy non-paper further proposes that the IGF establish links via ‘liaisons’ with relevant international negotiations on cyber security (Global Mechanism on Cyber Security), digital trade (WTO), human rights (UN Human Rights Council) and other digital issues, as well as with UN organisations such as UNCSTD, UNESCO and ITU. It is also suggested that the WSIS+20 pilot model of an Informal Multistakeholder Sounding Board (IMSB) be further developed in future intergovernmental digital negotiations.

Revision of the IGF modalities

The WSIS+20 Output Document called for a review of the IGF’s modalities. In January 2026, the IGF Secretariat issued a ‘Call for Intersessional Thematic Input’, which generated 61 proposals summarised in a 295-page document published in March 2026.[4] The "Intersessional" work is primarily to be carried out through the IGF’s sub-organisations. For instance, on 24 March 2026, the IGF Parliamentarian Track organised a conference in Montevideo entitled "Anticipating the Digital Future: Strengthening Parliamentary Governance and Democratic Innovation".[5] At the end of March 2026, a training course organised by the Judiciary Engagement Group for judges took place in Buenos Aires on the topic of "The Use of AI and Generative AI in Judicial Work".[6] Several Dynamic Coalitions have begun offering thematic webinars. The WG Strategy non-paper contains 49 proposals on the future programme development of the annual IGF, on inclusion and outreach, and on the working methods of the MAG.[7]

The IGF sub-organisations include

The Best Practice Forum (BPF) for cybersecurity;
The Policy Networks (PN) on artificial intelligence (PNAI), internet fragmentation (PNIF) and internet access;
The 32 thematic Dynamic Coalitions (DC);
The 138 National and Regional IGFs (NRI);
The Parliamentary Track;
The discussion platform for judges (Judiciary Engagement);
The Youth IGF

The IGF WG Strategy proposals on the working methods and composition of the IGF governance structures[8] focus on a more effective division of labour between the MAG and the IGF Secretariat. The MAG should concentrate on strategic considerations. Day-to-day work, including the evaluation of the nearly 1,000 proposals for IGF sessions, should be carried out by the Secretariat. The number of MAG members should be limited to 50.

As a guideline, no more than 40 per cent of MAG members should represent governments, with 20 per cent each representing civil society, the business sector and the technical community.
The renewal of the mandate for individual MAG members by the UN Secretary-General should be linked to an evaluation of their practical activity. This would allow for the early removal of inactive MAG members.
The rotation principle should be retained, with membership limited to a maximum of three years.
The MAG should primarily work virtually, but organise a hybrid meeting at least twice a year.
No financial allowances are provided for MAG members. Travel expenses may be reimbursed on a case-by-case basis.
The future of the Leadership Panel is still under discussion.

Unresolved issues regarding IGF funding

The financing of the IGF remains unresolved. The IGF Secretariat is awaiting the recommendations of UN Secretary-General Guterres, who is obliged under the OD to submit proposals by September 2026. At the WSIS+20 High-Level Meeting in December 2025, Germany announced that it would support the IGF with one million euros. In February 2026, State Secretary Richter from the BMDS was a guest of the IGF Secretariat in Geneva.

To-do list for the new MAG and planning for the 21st IGF

At its last (virtual) meeting on 23 March 2026, the 2025 MAG provided the new MAG with a long ‘to-do list’. The call for nominations for the 2026 MAG closed on 26 February 2026. The appeal on points of fact and law for the appointment of the new MAG members by UN Secretary-General Guterres is expected in April. By then, the venue and date of the 21st IGF are also to be announced. A webinar by the WG Strategy, "The post-WSIS+20 context: challenges and opportunities for the IGF", is scheduled for 2 April 2026. On 13 and 14 April, an IGF Retreat will take place in New York with a small group of selected stakeholders, the IGF Secretariat and UNDESA.

II. Artificial Intelligence

The debate on AI governance intensified and became more nuanced in the first quarter of 2026. New lines of conflict are emerging within an increasingly fragmented landscape of debate. The UN’s ongoing AI discussion process (AI Panel & AI Governance Dialogue) is being rejected by the US. The US is relying on its "Winning the Race: America’s AI Action Plan" (July 2025) and the "Pax Silica" initiative launched in December 2025, and is attempting to bring the so-called "Bletchley Process" over to its side. China has proposed the establishment of a "World Artificial Intelligence Cooperation Organisation" (WAICO)[9] based in Shanghai, is pursuing its "Global AI Governance Action Plan" from July 2025[10] and supports the UN process. India is seeking to capitalise on the successful "AI Impact Summit" in New Delhi (February 2026) and establish itself as a leading AI power in the Global South. The African Union (AU) is pursuing a "leapfrogging strategy" in which the younger generation, as developers of innovative AI applications, strengthens Africa’s AI sovereignty to prevent the emergence of neo-colonial dependencies.[11] Russia is prioritising AI innovations in the military sector. The EU continues to rely on the OECD-backed "Global Partnership on AI (GPAI)"[12], but also supports both the UN and the Bletchley Process and has somewhat relaxed the bureaucratic requirements of its AI law through a "Digital Omnibus".

Launch of the new UN AI bodies

The two new UN AI bodies (AI Panel & AI Governance Dialogue) began their work in the first quarter of 2026. On 3 February 2026, the 80th UN General Assembly appointed the 40 members of the "Independent International Scientific Panel on Artificial Intelligence".[13] Over 2,500 experts had applied. The resolution was adopted with 146 votes in favour, two against (USA & Paraguay) and two abstentions.

The establishment of the AI panel has got off to a bad start. The US justified its rejection by arguing that the UN should focus on its core mission – peace and security – "rather than attempt to regulate or stifle the development of cross-cutting and cutting-edge technologies that will determine economic and strategic competition in the twenty-first century".[14] Ukraine abstained because a Russian had become a member of the panel. The panel also includes two Germans: Bernhard Schölkopf, a professor at ETH Zurich, who is known for his opposition to autonomous weapon systems, and Maximilian Nickel, who is responsible for AI programmes at Meta in New York.
At the inaugural meeting on 2 March 2026 in New York, UN Secretary-General Guterres described the challenge facing the panel: "The world urgently needs a shared, global understanding of AI that is based not on ideology but on science, not on fake news but on knowledge. … Your task is to bring independent, credible science into the global debate – at a time when geopolitical tensions are rising, conflicts are raging, and the importance of safe and responsible AI could not be greater. In this fragmented context, an unbiased and trustworthy understanding of AI is essential … AI will shape peace and security, human rights and sustainable development. Your work will help decision-makers move from competing claims to shared facts, and from shared facts to workable solutions: establishing effective guardrails, harnessing innovation for the common good, and strengthening international cooperation."[15]
Spain, which is co-facilitating the AI Panel alongside Costa Rica, will host the panel’s first meeting from 22 to 24 April 2026. "AI may be unstoppable, but it cannot be ungovernable," said the Spanish representative at the UN General Assembly. The AI Panel must present an initial report by the end of June 2016, which will be discussed at the "Global Dialogue on AI Governance" on 6 and 7 July 2026 in Geneva. The work is being coordinated by the UN Office of New Technologies (ODET) under the leadership of Amandeep Gil Singh.[16]
On 18 March 2026, an initial informal consultation on the AI dialogue took place.[17] The co-facilitators here are the UN ambassadors from El Salvador and Estonia, Egriselda López and Rein Tammsaar. Five questions were discussed.[18] Among other things, the discussion focused on how to prevent fragmentation of the AI discussion landscape. Anna Osterlink from Article 19 called for a "holistic and multistakeholder approach" to AI governance and suggested that the UN’s AI Governance Dialogue should primarily have a coordinating function. One problem is the lack of a definition of AI governance. The lack of integration with internet governance was discussed. Apart from Nobel Peace Prize laureate Maria Ressa, who was co-chair of the IGF Leadership Panel, the panel did not include any experts from the internet community.

Bletchley Process: “AI Impact Summit” in New Delhi and the evidence dilemma

The ‘AI Impact Summit’ took place in New Delhi on 12 February. It is the fourth AI summit of the so-called Bletchley Process. In 2023, the then British Prime Minister Sunak had hosted the first AI summit in London. Attendees included numerous heads of government, amongst them US Vice-President Harris, French President Macron and German Chancellor Scholz, as well as the CEOs of leading AI companies from the US (including Sam Altmann of OpenAI and Mark Zuckerberg of Meta) and China. The "Bletchley Declaration"[19], adopted by consensus, sets out a series of principles for safe AI governance based on the 2019 OECD principles. However, at the subsequent conferences in Seoul (2024) and Paris (2025), the consensus began to crumble. At the Paris summit, US Vice-President Vance criticised the European AI Act and refused to sign a joint final document[20].

On 3 February 2026, shortly before the AI summit in New Delhi, the second ‘AI Safety Report’[21] was published. Following the Bletchley Conference, the EU, OECD and UN had agreed to form an expert group led by Prof. Yoshua Begion of the University of Montreal, the ‘AI Godfather’, to provide regular updates on AI safety. The 2026 report highlights an ‘evidence dilemma’ for policymakers. On the one hand, rushing to implement measures against potential AI risks before clear evidence is available could lead to ineffective or unnecessary regulations. On the other hand, a wait-and-see approach could leave society unprepared or make countermeasures impossible, allowing AI to escape human control. The report distinguishes between three security risks:
1. Malicious use: scams, cyber attacks, sexualised deepfakes, manipulation of the broader public's opinion, disinformation campaigns, etc.;
2. Malfunctions: unreliable conclusions, flawed generalisations, misdefined objectives, loss of control;
3. Systemic risks: dependence on a small number of AI monopolies, chain reactions in interconnected infrastructures, and the environmental costs of data centres with high energy and water consumption.
The AI Summit in New Delhi focused primarily on practical AI applications and the Global South. At the opening, India’s Prime Minister Modi said: "The fact that this summit is taking place in India is a source of pride not only for India but for the entire Global South." And he added: "Alongside the current generation, we must also consider what form of AI we will leave to future generations. The real question is not what AI can achieve in the future, but what we want to do with it today … AI is a transformative force: without direction, it leads to disruption; properly guided, it becomes the solution. How can we design AI to be human-centred rather than machine-centred, sensitive and responsible rather than reckless?"[22] UN Secretary-General Guterres called for "the future of AI not to be decided by a handful of countries or left to the whims of a few billionaires."[23]

The summit was attended by 35,000 experts from over 100 countries, including French President Macron, EU Commissioner Brunner, and the CEOs of numerous AI companies such as Sam Altman (OpenAI), Dario Amodei (Anthropic) and Sundar Pichai (Google). Civil society was sparsely represented.
Digital Minister Wildberger delivered a speech on behalf of Germany in the plenary session. Five German companies, including Bosch, Infineon and SAP, presented their AI applications in a German pavilion. A German-Indian AI pact was signed on the sidelines of the summit.[24]

The relatively brief "New Delhi AI Impact Summit Declaration"[25] avoids specific commitments and sets out seven very general principles (Chakras). This enabled 91 countries, including the US, the EU, China and Russia, to endorse it. Of particular note is the joint commitment to the multi-stakeholder principle and to the need for AI rules, which are, however, intended to be non-binding. "The AI Impact Summit will contribute to strengthening international cooperation and a multistakeholder approach to advance shared priorities as well as voluntary and non-binding guidelines and principles." Fifteen further documents[26] – which, however, were not signed by all participants – address topics such as AI and security, AI and climate, AI and democratisation, AI and child protection, etc., in more concrete terms. The "Framework for the Trusted AI Commons"[27], for example, has only 19 signatories, including Canada, Singapore, Egypt, France, Norway and Switzerland, but not Germany, the US, China or Russia. Similarly, the ‘Charter for the Democratic Diffusion of AI’[28] has only 23 signatories, including the Rockefeller Foundation and the Gates Foundation.
Speaking on behalf of the US, Michael Kratsios, Director of the Office of Science and Technology Policy at the White House, claimed global leadership in AI. "America is the birthplace of AI and the home of the frontier companies and hyperscalers," he said, adding, "the gold standard in AI is made in America."[29] He rejected the UN dialogue on AI governance, arguing that it fosters more fear than hope. "We must replace this fear with hope. Ideological, risk-oriented obsessions such as climate protection or social justice become pretexts for bureaucratic control and centralisation. In the name of security, they increase the risk that these tools will be misused for tyrannical control. We firmly reject global control through AI governance. We are convinced that the introduction of AI cannot lead to a better future if it is subject to bureaucracy and centralised control." The US is confident of winning the global "AI race". Other countries should join the "American AI Exports Program".

US strategy and the Anthropic case

In the US, the strategic importance that the Trump administration attaches to artificial intelligence is becoming increasingly clear. After US President Trump presented the general strategy under the title "Winning the AI Race"[30] in July 2025, in which he claimed global US leadership for international AI development, the US State Department organised a "Pax Silica Summit"[31] on 11 December 2025 and, together with a select group of countries – so-called "trusted partners" – launched a "strategic initiative to build a secure, prosperous and innovation-driven silicon supply chain – from critical minerals and energy resources to advanced manufacturing, semiconductors, AI infrastructure and logistics."

The initial signatories of the "Pax Silica Declaration"[32] are Japan, South Korea, Singapore, the Netherlands, the United Kingdom, Israel, the United Arab Emirates (UAE) and Australia. Observers were present from Taiwan, Canada, the EU and the OECD. India, Qatar, Sweden and Greece have since joined the initiative. On 26 March 2026, a Pax Silica Fund[33] was established, initially endowed with $250 million, to serve as a catalyst for private investment by US companies in AI projects in developing countries. [34]
On 7 March 2026, the White House published the long-awaited new US cybersecurity strategy (President Trump’s Cyberstrategy for America)[35], which fits seamlessly into the long-term AI strategy. "Cyberspace was born in America," the document states. From this, the claim to international leadership is derived: "The US’s cyber specialists and tools are the best in the world and can be deployed quickly and effectively to defend US interests … We will engage internationally through diplomacy, trade and operational measures to ensure that norms and standards reflect our values … We will streamline cyber regulation and reduce administrative burdens to ensure the private sector has the flexibility needed to keep pace with rapidly evolving threats." It is thanks to President Trump’s leadership that "a new era in cyberspace" has begun.
The announced "new era in cyberspace" is increasingly taking on a military dimension. The dispute between the AI firm Anthropic and the US Department of Defence plays a key role in this.

Anthropic had made the use of its AI contingent on the US government’s commitment not to deploy autonomous weapons that are beyond human control in the event of war, and to refrain from unwarranted AI surveillance of the US population.
On 22 January 2026, Anthropic had drawn up a constitution for its AI agent ‘Claude’.[36] Two of the four principles of ‘Claude’s New Constitution’ state: "Broadly safe: not undermining appropriate human mechanisms to oversee AI during the current phase of development; Broadly ethical: being honest, acting according to good values, and avoiding actions that are inappropriate, dangerous, or harmful.".
Defence Secretary Hegseth subsequently terminated all contracts with Anthropic and declared Anthropic a security risk. In his view, private companies had no right to interfere in decisions regarding military operations. Anthropic contested the termination of the contracts and was vindicated by a Californian court on 26 March 2026. The conflict raises a fundamental question: who has control over AI in the realm of national security: AI companies through internal security policies, or government agencies whose aim is to "win the AI race"?

China’s AI ambitions and WAICO

On 8 March 2026, China adopted its 15th Five-Year Plan, in which cybersecurity, internet governance and artificial intelligence play a prominent role.[37] According to the plan, China is set to become the world’s leading power in AI by 2030. The Five-Year Plan also calls for strengthening China’s international influence on the development of global AI governance. Chinese AI companies should increase their exports of applications, particularly to the Global South. [38]

In his speech at the World Economic Forum in Davos on 21 January 2026, China’s Vice-Premier He Lifeng reiterated the proposal to establish a World Organisation for AI Cooperation (WAICO) based in Shanghai. "China is ready to work with all other countries in a spirit of openness and cooperation to advance sci-tech innovation... and address global challenges in such areas as AI governance." [39]
A study by the Atlantic Council dated 20 February 2026 on the strategic adaptation of China’s Digital Silk Road (DSR) to the new AI challenges states: "The ‘Digital Silk Road’ has evolved into a key channel through which Beijing engages with countries of the Global South in the field of digital infrastructure and, increasingly, artificial intelligence (AI). Initially focused on ‘hard’ connectivity (e.g. telecommunications networks, fibre-optic and submarine cables, etc.), the Digital Silk Road has evolved since 2023 towards ‘soft connectivity’ – including AI governance diplomacy, standard-setting, capacity-building and service platforms. This shift reflects both China’s domestic economic imperatives (in particular, overcapacity in the high-tech sector and limited access to advanced markets) and its geopolitical objective of shaping a state-centred alternative digital order that strengthens state control over rights-based governance."[40]

Russia’s military AI focus

Russia has primarily geared its AI developments towards military applications. A study by the Centre for Strategic and International Studies (CSIS) from February 2026 identifies the development of an ‘Automated Command and Control System for Armed Forces and Weapons’ (ACCS) as the core of the digitalisation of Russian military doctrine. The ACCS is described as a network of integrated automated control and information systems, equipment complexes for automation, software-hardware complexes, and remote-controlled automated workstations distributed across the entire command hierarchy.[41]

On 17 December 2025, President Putin stated that the key areas for AI development must be "air and missile defence systems, command and control systems, electronic warfare capabilities, and unmanned aerial vehicles (UAVs)". "The Russian army must remain a technological leader. This means accelerating the introduction of robotics, information technologies and new materials into the armed forces, as well as expanding the use of AI in command and control systems and autonomous combat platforms." [42]
On 26 February 2026, President Putin signed an executive order establishing a new Presidential Commission on the Development of AI Technologies. Deputy Prime Minister Dmitry Grigorenko and Deputy Head of the Presidential Administration Maxim Oreshkin were appointed as co-chairs. The commission is tasked with coordinating both Russia’s domestic and international AI activities. According to a TASS report, it is intended to "ensure Russia’s technological leadership in the field of AI, including the creation and development of fundamental models and advanced services, computing capacities and the necessary electronic component base, as well as ensuring the power supply for such capacities."[43]

III. Cybersecurity

The most significant event in the global cybersecurity debate during the first quarter of 2026 was the inaugural meeting of the UN’s new "Global Mechanism" on 30 and 31 March 2026 in New York. A further round of negotiations on autonomous weapon systems took place from 2 to 5 March 2026 in Geneva. The topic of cyber security was also on the agenda at the Munich Security Conference from 22 to 24 February 2026.

Establishment of the “Global Mechanism”

The new permanent "Global Mechanism" (GM) on cybersecurity was established by the 80th UN General Assembly and seamlessly continues the work of the temporary "Open-Ended Working Group" (OEWG), founded in 2020, whose mandate had expired in December 2025. At the constituent meeting, agreement was reached on the governance structure of the new mechanism, the schedule for the next sessions and the formation of two thematic working groups.

Egriselda López, El Salvador’s ambassador to the UN, was elected chair of the GM. López is also co-facilitator of the Global Dialogue on AI Governance.
The first formal GM meeting will take place in New York from 21 to 26 July 2026. The meetings of the two working groups are also scheduled for December 2026 in New York. The two working groups (Dedicated Thematic Groups/DTGs) are to address security threats in cyberspace and capacity-building measures. However, the constituent meeting was unable to agree on either a concrete agenda for the first meeting or on the co-facilitators of the two DTGs.[44] This is now to be clarified in informal consultations ahead of the first meeting in July 2026.
The inclusion of non-state representatives was also discussed. Their capacity to make constructive contributions was recognised, but decisions should remain in the hands of governments. The procedures for the participation of non-state representatives are set out in UN Resolution 80/257 (Modalities on the Participation of Other Interested Parties and Stakeholders, including Businesses, Non-Governmental Organisations and Academia). NGOs that do not have ECOSOC status are subject to an accreditation procedure. If a government raises an objection, it must justify this objection. The Chair of the GM must then seek a mutually agreed solution through informal consultations. [45]

Negotiations on autonomous weapon systems (LAWS)

As expected, no significant progress was made during the negotiations on autonomous weapon systems (GGE LAWS) held in Geneva from 2 to 6 March 2026. Discussions centred on a number of additions to the ‘Rolling Text’ presented by the Chair of the GGE LAWS in February 2025. Accordingly, LAWS are now defined as follows: "A lethal autonomous weapon system can be characterised as a functionally integrated combination of one or more weapons and technological components, capable of identifying, selecting and engaging a target without human operator intervention in the execution of these tasks."[46].

The majority of states continue to press for the negotiations to be brought to a conclusion. In principle, there is a consensus on the ‘two-tier approach’. Under this approach, autonomous weapon systems that operate without human control should be banned. The use of AI in weapon systems under human control should be subject to the rules of international humanitarian law set out in the 1949 Geneva Conventions. However, the US, China, Russia, Israel and Ukraine in particular continue to block a final agreement.
The final meeting of the GGE LAWS is scheduled for early September 2026 in Geneva. The group’s final report will then be submitted to the next Review Conference of the Convention on Certain Conventional Weapons (CCW), scheduled for November 2026, under whose auspices the GGE LAWS operates. The UN General Assembly has expressed the expectation that a result will be available by the end of 2026.[47]

Munich Security Conference: A shift away from UN diplomacy towards offensive cyber strategies

Discussions at the Munich Security Conference included how to respond to cyber attacks in a changing global landscape. Numerous high-ranking experts called on Western states to respond more forcefully to attacks by third countries through offensive operations.

US Secretary of State Marco Rubio’s speech[48] was more assertive in tone than last year’s address by US Vice-President Vance, but contained similar content. Rubio reaffirmed the strong historical and cultural ties between Europe and the US, but also stated that the UN had failed and needed an alternative. He added that the US and the EU had made historic mistakes in recent years that now needed to be corrected. The US would prefer to move forward together with a strong Europe. However, the US could also go it alone. Sean Cairncross, Cyber Director at the White House, elaborated on this: "US first doesn’t mean US alone".[49] However, Europe must rethink its digital rules. Regulation has been secondary and has lagged behind digital developments. Innovation must come first. EU Commissioner Henna Virkkunen explained that the EU’s ‘Digital Omnibus’, which aims to reduce regulatory bureaucracy in the digital sector, does not mean digital deregulation. [50]
During a panel discussion on cyber diplomacy[51], efforts to establish common rules at a global level – such as the UN Convention on Cybercrime or the ‘Global UN Mechanism’ for cybersecurity – were viewed as largely peripheral and having little impact on the new generation of digital warfare. The discussion centred less on "norm-making" or "norm implementation" and more on "norm circumvention". Referring to potential adversaries’ disregard for international law, British General Magowan said: "If we stick to the rules, we will lose." He called on European heads of state and government to be more willing to take risks. However, he also made it clear that taking more risks did not mean ignoring international law; rather, international law should be interpreted creatively. According to BND chief Martin Jäger, the West must rethink its options regarding cyber attacks. It is not enough to condemn cyber attacks from abroad. He proposed asymmetric offensive responses: "If the Russians attack our rail network’s data centre, we should not attack their rail system, but could withdraw cryptocurrency from their accounts."
As the problem of attributing cyber attacks remains complicated and grey areas exist, direct counterattacks increase the risk of escalation with cascading effects that can spiral out of control. The problem is that cyber defence is expensive, whereas cyber attacks are cheap. Hostile states can deploy private criminal intermediaries (proxies) to destabilise the political landscape of their adversary. "Cybercrime as a service" is not only economically attractive to organised crime, but can also strain relations between states, regardless of UN conventions.

IV. Global negotiations on the digital economy

For a quarter of a century, two issues have been on the agenda in global negotiations on the digital economy: tariffs on digital services and a digital tax. In the 2010s, the topic of ‘data governance’ was added to the agenda. Tariffs are being negotiated within the World Trade Organisation (WTO). The digital tax is the subject of negotiations within the OECD and the G20 and was partially regulated in the 2021 BEPS Agreement. Work has been underway since 2024 on a "UN Framework Convention on International Tax". Data governance has been a topic of discussion at UNCTAD since the 2010s and was taken up by the Global Digital Compact in 2024. This is now being addressed by a UNCSTD Working Group on Data Governance (WGDG).

WTO: e-commerce agreement and global customs moratorium

When international electronic commerce (e-commerce) expanded significantly in the 1990s, the World Trade Organisation (WTO) decided in 1998 not to impose tariffs on cross-border electronic services for the time being.

The moratorium was temporary but was repeatedly extended. In the 2010s, a number of developing countries (India, Indonesia) called for an end to the moratorium and for an agreement on international data traffic (Digital Trade Pact) regulating the levying of taxes on digital services. They hoped this would generate billions in additional revenue.
This met with opposition from the US and the majority of OECD countries, which, for their part, wanted to remove the time limit on the moratorium and make it permanent. The majority of WTO members endorsed a corresponding ‘Joint Statement Initiative (JSI) on eCommerce’, which was launched on the sidelines of the World Economic Forum in Davos in 2019. The aim was to enshrine the moratorium in a permanent agreement. However, negotiations progressed only slowly, partly because the majority of WTO members were seeking a more comprehensive e-commerce agreement that went beyond a mere extension of the moratorium, which was rejected by some OECD countries, and above all the US.
After no agreement was reached at either the 12th (2022 in Geneva) or the 13th WTO Ministerial Conference (2024 in Abu Dhabi), the negotiating group led by Singapore, Japan and Australia was able to agree on a consolidated text for an "Agreement on Electronic Commerce"[52].

The agreement establishes uniform rules for cross-border e-commerce. Invoices, signatures, contracts and commercial documents in electronic form are recognised as equivalent to their paper-based counterparts. It prohibits tariffs on digital services and requires the contracting parties to establish a legal framework that protects online consumers from misleading and fraudulent business practices. Cybersecurity in electronic transactions is to be strengthened. It is also intended to promote the "digital inclusion" of the least developed countries.
In Yaoundé on 26 March 2026, the agreement was signed by an initial 65 governments out of the total 166 WTO members, which, however, account for 70% of global e-commerce. It is a so-called "plurilateral agreement" that is not yet part of the general WTO regulations. The agreement will enter into force for the signatory states once the 45th instrument of ratification has been deposited. These states will then no longer be able to levy customs duties on cross-border e-commerce.
However, extension of the moratorium binding on all WTO members failed. Consequently, states that have not acceded to the agreement may now levy tariffs on, for example, streaming content, software downloads and other electronic products. The US wanted only a permanent extension of the moratorium and rejected the e-commerce agreement. Brazil and Turkey did not agree with this. They made their approval of an extension of the moratorium conditional on concessions from the US, which the US did not make.
Following the Ministerial Conference, the WTO Council in Geneva has now been tasked with finding a compromise. The US was dissatisfied with the outcome, and US Ambassador Greer expressed scepticism that an agreement could be reached retrospectively. "This organisation will play only a limited role in future global trade policy efforts," he said at the conclusion of the conference.[53] He announced that the US would in future regulate its trade policy through bilateral agreements. The EU favours a dual strategy: it is expanding the network of bilateral and plurilateral trade agreements, for example with Latin America, India, Canada and Australia, but remains interested in strengthening the WTO.[54]

Global digital tax and BEPS compromise

Global regulation of a digital tax has been the subject of negotiations since the 2010s within a joint OECD-G20 working group, which reached an agreement in 2021 on measures to combat base erosion and profit shifting (BEPS). 141 countries have now acceded to this BEPS agreement. The core element of Pillar 2 of the BEPS agreement is an effective minimum tax rate designed to reduce opportunities for profit erosion and profit shifting. The global minimum rate has been set at 15%.

The US had withdrawn from the negotiations under the first Trump administration in 2019. The Biden administration rejoined the negotiations in 2021 and also signed the BEPS agreement. However, it has not been ratified by the US Congress.
Following President Trump’s return to the White House, the US government declared in February 2025 that it would not accept extraterritorial taxes on American tech companies and threatened sanctions. On 28 June 2025, the finance ministers of the G7 group agreed on a compromise that preserves the fundamental elements of the BEPS agreement but allows for exemptions for US corporations. These were to be enshrined in a so-called ‘Side-by-Side Agreement’ (SdS). The "G7 Statement on Global Minimal Taxes" of June 2025 states: "A side-by-side system would fully exclude U.S.-parented groups from the UTPR and the IIR in respect of both their domestic and foreign profits."[55] As the OECD announced on 5 January 2026, these exemptions were enshrined in a so-called "Side-by-Side Package". [56]

Progress on the UN Framework Convention on Tax Cooperation

Independently of the OECD/G20 negotiations, discussions are taking place under the auspices of the UN regarding a UN Framework Convention on International Cooperation in Tax Matters.

On 24 December 2024, the UN General Assembly adopted UN Resolution 79/235, establishing a "Member State-led, open-ended intergovernmental negotiating committee (ING) for the purpose of drafting the United Nations Framework Convention on International Tax Cooperation".[57] The ING’s negotiations began in August 2025 in New York.
The 4th ING session took place in New York from 2 to 13 February 2026. The subject was a first draft convention comprising 28 articles. Article 5 of the draft, for example, establishes the right to levy taxes where value is generated: "The Contracting States agree that all jurisdictions in which value creation, markets, revenues or economic activities take place have the right to tax a portion of the income derived from these activities."[58]
The 5th and 6th rounds of ING negotiations are scheduled for August 2026 in New York and December 2026 in Nairobi. The aim is to conclude the negotiations by December 2027.[59]

UNCSTD report on data governance

Data governance has been the subject of discussions within UNCTAD for years. In its annual "Digital Economy Report", UNCTAD has for years highlighted the need for international regulation on cross-border data flows. The 2021 Digital Economy Report, for example, refers "to the complexity of cross-border data and data flow regulation in relation to sustainable development." It also emphasises "that the international debate on the regulation of cross-border data flows is at an impasse and that positions tend to be polarised. The current regulatory landscape is patchy and reflects widely differing approaches across countries, which are significantly influenced by the major economic powers. An international framework is urgently needed to address this situation."[60]

The issue was raised during the negotiations on the ‘Global Digital Compact’. Article 48 of the GDC (September 2024) provides for the establishment of a UNCSTD Working Group on Data Governance (WGDG). The WGDG is tasked with developing recommendations "towards equitable and interoperable data governance arrangements, which may include fundamental principles of data governance at all levels as relevant for development; proposals to support interoperability between national, regional , and international data systems; considerations of sharing the benefits of data; and options to facilitate safe, secure, and trusted data flows, including cross-border data flows as relevant for development."[61].
The first meeting of the WGDG took place on 1 May 2025 in Geneva. It comprises 54 members: 27 governments – including the German Federal Government – and 27 non-governmental representatives – including Michèle Finck from the University of Tübingen. Peter Major (Hungary), Muhammadou M.O. Kah (Gambia) and Claire Melamed (civil society) were elected as co-chairs. Peter Major passed away in autumn 2025 on the eve of the 3rd WGDG meeting.
The 5th session of the WGDG took place on 2 and 3 March 2026 in Geneva.[62] The agenda comprised four key areas: 1. Data Governance Principles, 2. Interoperability between national, regional, and international data systems, 3. Sharing the Benefits of Data, and 4. Data Flows. Extensive working papers were produced on all four topics. These form the basis for the report to be submitted to the 81st UN General Assembly in autumn 2026.

China’s new ‘World Data Organisation’ (WDO)

On 31 March 2026, a new "World Data Organisation" (WDO) was founded in Beijing. The WDO is conceived as a non-governmental and gemeinnützige internationale Organisation, which describes itself as "a platform for global data collaboration that is practice-oriented, industry-focused and designed for multi-stakeholder dialogue and cooperation"[63]. The founding event was attended by 200 participants from 40 countries. A charter for the new organisation and a list of its founding members have not been published.

In a message of greeting to the inaugural event, Chinese President Xi Jinping wrote that the WDO can help "to bridge the data divide, to unlock data’s value and to power the digital economy. The WDO provides a beneficial platform for deepening international data cooperation and improving global data governance".[64] China Daily wrote: "Unlike treaty-based institutions such as the WTO, the WDO will not establish binding rules. Instead, it will focus on coordinating policy measures, recommending standards, technical cooperation and capacity building, and will act as a ‘trust anchor’ for cross-border data flows."[65] China Daily points out that whilst the WDO is legally registered in China, it is to be structured as a voluntary international association with members from various countries and sectors.[66]
Among those attending the founding conference was Muhammadou Kao, Co-Chair of the UNCSTD WGDG. The WDO is based in Beijing. The WDO is set to receive recognition from the UN by 2030. Yang Jie has been appointed Secretary-General of the WDO. Yang Jie was Chairman and Party Secretary of China Mobile Communications Group Co. Ltd. He is currently a member of the Economic Committee of the National People’s Congress.[67] Tan Tieniu, President of Nanjing University, was elected Chairman of the WDO Council. Members of the WDO Council include Wilton Mota, CEO of the Brazilian company Serpro[68], a provider of digital services.[69] No further details are currently available regarding other members of the WDO Council.