UN Norms: Tackling the Rise of Cyber Capabilities

The conclusion of the UN’s Open Ended Working Group (OEWG) – the process responsible for devising ‘rules of the road’ for responsible state behaviour in cyberspace – finished its work in July 2025, marking the end of a cycle of negotiations on cybersecurity begun 21 years ago. What will take its place remains uncertain and raises issues regarding the future of norms and rules for states in cyberspace.

The Final Report of the OEWG broke little new ground because states did not wish to go beyond the ideas discussed in six earlier Groups of Government Experts (GGE) and two OEWGs. This means the substantive agreements in the 2015 GGE – when states agreed on 11 norms – are the high-water mark of UN cyber negotiations. But it has been over a decade since that agreement and further progress requires states must now transition from the OEWG to a new ‘Permanent Mechanism.’

How Did We Get Here?

GGEs are a UN mechanism to study difficult topics and make recommendations to the Secretary General. The first GGE in 2004 failed to reach consensus because of US opposition. The second GGE (2009-2010) agreed on a brief work program to develop norms, confidence building measures (CBMs) and (at the insistence of the South African expert) capacity building. The 2013 GGE amplified these topics and the 2015 GGE reached the landmark agreement on 11 norms. The 2015 GGE, held during a time of fading international comity, created a global framework for responsible state behaviour in cyberspace.

To continue the work of the GGEs, the General Assembly created an OEWG (the UN uses OEWGs to discuss major issues and involve all member states) to further develop the 2015 GGE’s work. There have been two cyber OEWGS, the first from 2019-2021 and the second from 2021-2025. The second OEWG’s Final Report had four noteworthy results. It solidified the place of the 11 norms first agreed in the 2015 GGE to define responsible state behaviour. It reinforced the applicability of international law to cyberspace. It established regular institutional dialogue (called the ‘Global Mechanism on developments in the field of ICTs in the context of international security and advancing responsible State behaviour in the use of ICTs’). It made capacity building central to UN cyber efforts.

The report lays the foundation for future work on responsible state behaviour using capacity-building and the Global Mechanism. But the GGEs began at the dawn of global connectivity. Their success in placing cyberspace in the context of existing agreements (principally the UN Charter and the corpus of international law) was an essential first step and anchors cybersecurity squarely in the framework of state practice and international relations. The 2021-2025 OEWG usefully continued this and widened the discussion to the entire UN membership. Now, new ideas are needed to strengthen stability and reduce the chance of greater conflict.

Precedents and Obstacles for Future Talks

An adequate history of UN cyber negotiations has yet to be written. Negotiations began with a Russian proposal in 1998 to create a binding legal agreement for cybersecurity. The proposal did not achieve consensus. Russia then proposed the creation of a GGE under the auspices of the First Committee (Disarmament and International Security). This now hampers progress.

If the major problems for cybersecurity are crime, espionage and warfare, the GGEs and OEWG did not address them. Cybercrime falls under the purview of the UN Office on Drugs and Crime, not the First Committee and member states have already agreed to a binding Cybercrime Convention (also proposed by Russia). UN negotiations routinely exclude espionage from discussion in a reflection of the political difficulties of the topic. The GGEs and OEWG dealt with offensive cyber action only in general terms and agreed to excluded privacy and data protection on the grounds these are not international security issues (although every Report acknowledges the importance of human rights). In effect, the most important challenges to peace in cyberspace were taken off the table for negotiation. This allowed for agreement, but constrained OEWG outcomes and will make it difficult for any successor to address major security problems.

The Reality of State Interests

The exclusion of these topics reflects the views of major cyber powers to limit discussion of cyberwar and espionage. The 2015 text contains a few references to the rules governing armed conflict, but Russia and China argued (with support from other nations) direct references to the Laws of Armed Conflict (LOAC) would legitimize conflict in cyberspace, which they argued should be a ‘zone of peace.’ The 2025 OEWG saw a similar debate over ‘exclusive peaceful use.’ While the concept (derived from nuclear weapons free zones) is absurd, especially coming from Russia and China, not all Western powers were unhappy with this impasse, as it did not constrain their freedom of action in cyberspace curtailed beyond the limits on the use of force already found in IHL. Cyber war is closely linked to espionage, further complicating the discussion as major actors refuse to engage on this sensitive topic as well. In effect, no major power is willing to renounce offensive cyber operations. This limited and shaped the GGE and OEWG discussions.

Cyberspace also has unique features that shape negotiation. Its technologies are not exclusively military nor do they belong solely to states. Cybersecurity unavoidably involves the private sector. Defining an appropriate role for the private sector role has always been an issue in the GGE/OEWG process. At first, states unanimously opposed private sector participation - at one famous dinner in 2013, organized by a large American technology company and held at the German UN Mission, the host turned to the assembled GGE negotiators after some introductory remarks and asked if any wished to speak. There was complete silence and the dinner ended shortly thereafter. Silence would be unlikely today, but the fundamental objections to non-state participation have not changed.

Broadening participation works against meaningful agreement. Nations with advanced cyber capabilities are unwilling to engage with outsiders on a fundamental aspect of state power. Two major cyber powers will not engage seriously if NGOs and companies are involved. The Chinese and the Russians in particular resist the involvement of private sector and non-governmental entities. Some countries maintain complete control of private sector actors and see them as mere appendages of the state.

The exception has been groups accredited with the UN’s Economic and Social Council (ECOSOC), which are already accredited. Others can be blocked without explanation by any UN member State. Participants have said the result is overrepresentation by groups with a human rights or development-oriented agenda. Nor do the normative agendas of some non-governmental groups have traction with states. This is an unfortunate product of political reality and dilutes the chances of meaningful discussion.

The new Mechanism is open to all member states, regardless of their capabilities or interests in cyberspace. This follows the OEWG precedent, which could be an obstacle to future agreement. Like the GGEs and OEWGs, decision-making is complicated by the need for consensus. And there are a few states, too large to be easily ignored, whose diplomats often play the role of ‘spoilers.’ Unlike previous major arms control initiatives, the leading belligerents have not engaged in regular security dialogues (much less negotiations) in over a decade and the Mechanism is not a substitute.

Most importantly, cyber operations do not pose an existential threat. This removes the incentives for serious, sustained talks. In the Cuban Missile Crisis, the US and Soviets narrowly avoided nuclear war. This led them to begin a long series of interlocking negotiations on nuclear weapons, space, conventional forces and weapons of mass destruction. The risk from cyber-attacks has not prompted similar, sustained, high level negotiations. If the annual UNGA speeches by heads of state are an indicator, cyber is something to be noted and deplored but not linked to any specific proposals for action.

Are New Confidence Building Measures Useful?

That lack of urgency limits the usefulness of CBMs. The 2025 OEWG made progress on CBMs, but with important limitations. These are clearer if we consider the precedents for cyber CBMs, which came from strategic arms control. The 2010 and 2013 GGE language on CBMs drew from both the Conventional Forces in Europe (CFE) Agreement and the Helsinki Final Act, used CBMs to reassure potential opponents and prevent surprise attack. These precedents appealed to the 2010 Russian GGE Chair, as he had deep experience with strategic arms control (and Russia, as the proposer of the GGE and OEWG processes, felt a degree of ownership tacitly recognized by the UN’s Office of Disarmament Affairs, sponsor of the GGE).

The difficulty of developing similarly concrete CBMs for cyber actions is a major obstacle. Crucial elements of traditional CBMs, such as prior notification, thresholds, numeric ceilings and geographic limits, are not in the OEWG Report because it is difficult to see how to apply them to the intangible domain of cyberspace. One of the most important elements of traditional CBMs – prior notification of potentially destabilizing acts – is especially inapplicable for cyber actions, which rely on covertness and surprise.

Additionally, while the drafting of the first GGE reports drew on CFE and Helsinki, neither of these agreements involved the UN. They were binding agreements among belligerent major powers and their allies. The international security environment is no longer a Cold War bilateral contest. The rise of new powers who demand a greater voice in defining the international order complicates discussion in ways not conducive to agreement.

Also, the CBM exchange of information is largely among non-belligerents (and this is also true for all but one of the regional CBMS agreements). Information sharing among non-belligerents does not change the calculus of risk for international peace. While regional CBMS have won acceptance, CBMs among all UN Member States have limited benefit for stability, since most countries are not potential cyber belligerents. The effect is to make the OEWG CBMs largely symbolic and limit their ability to contribute to international peace and stability.

The chief weakness of the OEWG’s cyber CBMs is they apply to an arena where few states have the capability for hostile action yet they lack the mechanisms for states with those capabilities to interact directly. Before 1990, if troops, tanks, or aircraft had crossed the East-West Border, there would have been a prompt and forceful response, accompanied by high-level diplomatic protests and warnings. Only a handful of cyber incidents have produced a similar reaction. Cold War CBMs reduced the risk of miscalculation or that an incident was the prelude to general war, and this created incentives for engagement. These incentives are now lacking, in part because of the absence of an existential threat from cyber-attack and in part because of the dilution produced by universal participation.

The Global Mechanism and Risk

Some point to the Global Mechanism itself as a CBM. There is some truth to this, depending on how active and energetic the Mechanism proves to be. The Global Mechanism will be subsidiary to the General Assembly and report to the First Committee. It will hold annual plenary sessions and has two subgroups, one on emerging threats, norms, international law, CBMs) and another specifically on capacity building. The concern expressed by many is it will turn into the cyber equivalent of the Committee on the Peaceful Uses of Outer Space (COPUOS), a well-intentioned body famous for its inaction.

The conclusion of the OEWG is the conclusion of the sequence of arms negotiations begun when the Cuban Missile Crisis frightened the major belligerents to the point they began talks to reduce risk. This is not the case now. There is no clear path for next steps in international security writ large, much less international cybersecurity. No nation has put forward at a senior level new ideas for a compelling agenda for the multilateral discussion of cybersecurity. Progress will not be possible for the next few years, but the need for multilateral agreement on governance and security in the digital environment will continue to grow. The OEWG, to use a sports metaphor, brought more players into the field and kept the ball in play – and may be its most valuable contribution. The Global Mechanism can build on this by laying the foundation for future progress.

An Opportunity to Have Tough (Yet Productive) Conversations

A Global Mechanism is here to stay with no term limit to its mandate. States have an opportunity at hand – and it is more than one of setting with ‘dialogue is better than no progress’ in a deeply protracted geopolitical environment. As a permanent mechanism, states should not shy away from addressing tough topics. The question is, which ones to address first and how to properly tailor the discussions in a constructive manner.

The Mechanism has two groups, one on specific cybersecurity challenges, where the OEWG conclusions are discussed as practical tools rather than abstract concepts, and another on capacity building. These allow countries to share experiences and address specific challenges in implementing the OEWG framework, particularly those for cyber resilience and defining capacity-building needs. Both groups could be used to discuss hard issues and build common understandings in a fluid and dynamic technology space.

Change in national policies create an opportunity for an expanded agenda. Perhaps as many as fifty countries have offensive cyber capabilities and some, from Japan to NATO, are reorienting their policies from deterrence to ‘active cyber defence’ – and are speaking publicly about it as well. Rather than join in a tempting scepticism towards multilateral institutions, the agreed norms and the Global Mechanism provide an opportunity to address problems arising from the coercive use of cyber tools. Offensive cyber operations and cyber espionage pose the greatest risk to international peace and stability from cyberspace. The topics have not been part of the previous negotiation, but progress requires putting them back in play even though they will be difficult, serious negotiations do not provide immediate gratification, and both are topics the Global Mechanism can begin to address.

This discussion will take years, but that is to be expected for serious topics and meaningful commitments. The OEWG agreement built on the wok of 21 years. And adding new norms is not a simple task. There have been previous attempts to expand the norms agreed in 2015, but none of the proposed additions were compelling. A Commission funded by the Netherlands suggested new norms, but these were either marginal or redundant. Two new norms were proposed in the OEWG discussions. They focused on access to information technology and on countermeasures. Neither was adopted.

This does not mean the existing 11 norms are adequate; it means expanding or refining them will be difficult, given the sensitivity of the most salient topics. The First Committee (whose agenda is often driven by the major powers) lacks the mandate and the intellectual groundwork to address topics like espionage and warfare, nor are powerful states willing to give it this mandate. Agreement on new norms must wait for a different political environment, but their discussion can begin now.

Moreover, the 11 norms could use a technology ‘refresh.’ They are a product of their times and shaped by the cyber technology at that time. The current norms do not address later developments like cloud computing, big data and artificial intelligence. This means proposals to update existing norms through discussion in the Global Mechanism would be valuable.

A Norm on Offensive Operations

Existing commitments to refrain from the use or threat to use force (put aside for a moment how to define force in cyberspace) and to respect sovereignty and political independence, provide the basis for additional normative constraints on State behaviour. Offensive cyber operations are to some extent already covered by the UN Charter’s Articles 2/4 and 51 and by other protections the Charter accords to member states. While states will not agree to a definition on the use of force in cyberspace, (major states regard it as a political decision they wish to reserve to themselves), they could adopt cyber offense as a topic for discussion in the Global Mechanism.

In addition, an opportunity is provided by paragraph 29(d) of the 2015 GGE Report, which ‘notes the established international legal principles, including, where applicable, the principles of humanity, necessity, proportionality and distinction.’ These principles were taken directly from the Law of Armed Conflict (LOAC) and International Humanitarian Law (IHL). Russia and China objected to them appearing by name in the Norms section of the Report but allowed them to appear in the Section on International Law (without reference to their LOAC origins). China’s position on the applicability is now more nuanced and more open to discussion of international law. An ambitious effort in the Mechanism could build on the 2015 language.

A new norm could make existing UN Charter commitments more explicit in their application to offensive cyber operations. A salient example involves placing disruptive malware on critical infrastructure networks. A new norm could say, for example, States agree to refrain in peacetime from placing malware on other nation’s infrastructure for later use in disrupting operations. While the absence of prepositioning will create operational difficulties for an attacker (and no nation will renounce surprise attack) and while ‘peacetime’ is itself a term subject to qualification, agreement to not preposition in peacetime would provide a stabilizing effect.

As part of this, States could usefully expand CBMs to reinforce norms on the use of force. These could include explicit exchanges of information offensive cyber capabilities and, given sensitivities, develop proxy measures, such as the size of cyber forces), doctrines for military use and (in general terms), cyber budgets. Other categories of exchangeable data could be identified to expand transparency on coercive capabilities.

Norms on Cyber Espionage

Reaching consensus on norms constraining cyber espionage will be exceptionally difficult, but one potential avenue is to build on the Charter provisions that apply to respect for sovereignty. Such an agreement would have implications for protecting privacy and data sovereignty, topics that have become more important for security and will continue to increase in importance as AI creates new demands and uses for data. Technological change has introduced new risks that are not yet on the First Committee’s agenda.

While practice and law accept the state use of espionage, there is one precedent for a norm, the short-lived 2015 US-China agreement on a norm against certain types of state-sponsored cyber espionage. That agreement involved a mutual pledge neither government ‘will conduct or knowingly support cyber-enabled theft of intellectual property, including trade secrets or other confidential business information, with the intent of providing commercial advantage to their companies or commercial sectors.’ It implicitly left room for both countries to continue cyber espionage for traditional national security purposes. The agreement was endorsed by the leaders of the G-20, the world’s richest economies. The G-20 Communique stated ‘no country should conduct or support ICT-enabled theft of intellectual property, including trade secrets or other confidential business information, with the intent of providing competitive advantages to companies or commercial sectors.’

The agreement itself collapsed after a few months as Chinese entities (private and state) resumed commercial cyber espionage without objection from the US. Several issues led to the collapse, in particular differences between China and the US on when commercial espionage was justified by national security. Both the US and China agreed espionage on commercial entities for national security purposes was justified, but the Chinese argued stealing purely commercial technology to build their technology base was also justified by national security. IP theft by private Chinese actors from both their foreign and Chinese competitors was something the Chinese government was unable or reluctant to control.

Despite this, the elements of the 2015 agreement could be developed into a new norm, where UN member states would commit not to use government agencies or proxies to use cyber means to steal IP for commercial benefits. This introduces the topic of espionage in the context of commercial cybercrime, making it more amenable to discussion. It puts the topic of espionage into play and addresses one of the significant weaknesses of the OEWG, which is that it does not directly address the major threats to peace and stability in cyberspace. Immediate agreement is unlikely, but the 11 norms themselves took years of hard work and a failure to address hard topics will make the Mechanism feeble or irrelevant.

Reinforcing Accountability

A lack of accountability is the chief problem for cyber norms. Nor is there any agreed mechanism for verification. The OEWG Chair attempted to address this by identifying voluntary measures for norms implementation in a ‘Voluntary Checklist of Practical Actions for the implementation of voluntary, non-binding norms of responsible State behaviour in the use of ICTs.’ Even if adopted by all nations, the Checklist set a low baseline of minimal requirements, such as acknowledging receipt of a message. There is ample space for further work. 

Accountability for new norms on offensive cyber operations and cyber espionage faces the difficulties of verification that apply to most malicious actions in cyberspace. Third party verification (like an IAEA for cyber) is technically difficult and politically unacceptable. This means the effect of a new norm, as with the existing norms, would be a commitment to principles even if these are often observed in their breach. Violations of national sovereignty in cyberspace are routine and not immediately detectable. Nations have become inured to them. In the long term, however, norms create a framework for stability.

In the UN, accountability is strengthened when an issue is referred to the Security Council for action. Finding ways to make greater use of the Security Council would improve cyber accountability. One precedent comes from proliferation-related agreements like the Missile Technology Control Regime (MTCR), a nonbinding agreement among major states to refrain from certain behaviours deemed to be irresponsible. The MTCR relied on voluntary compliance and verification was provided through shared intelligence and national technical means among the member states. Since MTCR members shared a common view of the risk of missile proliferation, there have been very few instances when norms were transgressed. Transparency and shared objectives create accountability. The Permanent Mechanism could usefully reinforce both, and reinforce the direct involvement the Security Council – the only body whose decisions member states are then obligated to implement.

Difficulty is Not an Excuse for Inaction

Espionage and warfare are difficult areas to address, but if the Global Mechanism confines itself to reiterating the 2015 Report or to an anodyne listing of threats, it will have the same lack of salience (in terms of enhancing international stability) as the Committee on the Peaceful Uses of Outer Space (COPUOS). While this may not change until there is a greater sense of risk from malicious cyber action, the Global Mechanism could usefully provide a venue to begin discussion, familiarize member states, and develop common understandings. Reaching agreements in the UN on complex and sensitive issues take years to negotiate. States move cautiously when making agreements that affect their core interests. Expectations for work programs and progress should take this into account.

Russia’s original goal in proposing a binding cyber agreement was to constrain the US in developing for military use a technology where it (and others) was far behind. A treaty was unattainable in 1998 and is even less attainable today. The idea of a binding accord produces a neuralgic reaction from the US and its allies in some measure because as a Russian proposal, making it immediately suspect. In any case, negotiations would be premature, given the continued pace of technology development. It would be also challenging given the Russian desire to use agreement to constrain speech.

Given the current political landscape and the difficulties of verification, there is now no space for any new UN agreement on norms. Until that changes, the Global Mechanism could become a venue for building knowledge and consensus.

The first GGEs took place when the internet was new and unfamiliar. The GGEs ‘normalized’ cyberspace by placing it in the context of existing state practice and agreement. The OEWG widened the discussion to involve all UN member States and reinforced the commitments on the implementation of norms, capacity building and the application of international law. The Global Mechanism ow has an opportunity to further extend this. Immediate progress is unlikely, but the foundation can be laid now for future agreement.

© James A. Lewis, 2026, published by RUSI with permission of the author.

The views expressed in this Commentary are the author's, and do not represent those of RUSI or any other institution.

For terms of use, see Website Terms and Conditions of Use.

Have an idea for a Commentary you'd like to write for us? Send a short pitch to commentaries@rusi.org and we'll get back to you if it fits into our research interests. View full guidelines for contributors.