The Weaponisation of Cyberspace and a New “Global Mechanism” at the UN
Blog post by Amb. Paul Meyer, Senior Advisor of ICT4Peace.
On 11 July 2025, at its Eleventh substantive session, the UN Open-ended working group on security of and in the use of information and communications technologies (2021-2025) adopted in New York adopted its Draft Final Report, as submitted by its Chair.
See the letter of the Chair of the OEWG and the adopted draft Final Report here
And find the the blog post by Amb. Paul Meyer here. commenting on the proposed new “Global Mechanism”:
“It is easy to be disillusioned over the current state of cyberspace. A
promising realm for information exchange has become a
treacherous environment, full of threats originating with state and
non-state actors alike. It has become increasingly difficult for the
average user to safely navigate the Internet, while even for states
the volume and sophistication of offensive cyber operations pose
major risks for their security.
To date, some forty states have established cyber units within their armed
forces, most of which are considered capable of “offensive” cyber
operations. With the establishment of CAFCYBERCOM Canada has
jettisoned the euphemistic “active cyber defence” terminology and now
acknowledges it has developed “offensive cyber operations capabilities”.
Alongside the military there are also intelligence agencies who have
invested heavily in cyber espionage. A challenge for victims of cyber
intrusion is to determine whether the operation is intended to extract
information or to destroy systems and disrupt operations. Unlike other
forms of military action, offensive cyber operations are carried out covertly
and states rarely acknowledge that they have been responsible for an
attack on another state. Even the high profile Stuxnet attack in 2009
against Iranian nuclear facilities has never been officially claimed by the
presumed perpetrators – the United States and Israel.
On a daily basis, we learn of state-conducted cyber operations that have
exfiltrated tetra bytes of data or compromised critical infrastructure even
as those responsible hide behind a cloak of plausible deniability. China
tops the list of adversarial cyber actors according to the latest threat
assessment by the Canadian Government, stating that “Over the past four
years, at least 20 networks associated with Government of Canada
agencies and departments have been compromised by PRC cyber threat
actors”.
Alongside this real-world use – or abuse – of the Internet there have been
efforts to subject this novel means for projecting power to some form of
constraint. The weaponization of cyberspace can negate the societal and
developmental benefits of this environment and “rules of the road” need
to be agreed to curb destructive behaviour.
The United Nations has been a key venue in the effort to regulate state
conduct in the quintessentially universal realm of cyber space. Since
1998, the UN General Assembly has had an agenda item on
“Developments in the Field of Information and Communication
Technologies in the context of International Security”. Over 27 years, the
UN has had a series of processes to consider state cyber conduct. The
most significant output from this work was a 2015 document that
enumerated eleven voluntary norms of “responsible state behaviour in
cyberspace” that was subsequently endorsed by a consensus General
Assembly resolution. Amongst these norms were agreement to refrain
from cyber attacks directed at critical infrastructure on which the public
depends, to prohibit attacks against Computer Emergency Response
Teams (the “first responders” to cyber incidents) and to ban the use of
proxies. This document is referred to as the “normative framework” and
represents a high-water mark of international cooperation regarding state
cyber behaviour.
The other major outcome of this lengthy process of diplomatic
engagement was agreement this July on a final substantive report of an
Open-Ended Working Group on cyber security after four years of
proceedings. Importantly this report establishes a mechanism for on-
going consideration at the UN of cyber security issues. This “Global
Mechanism” would entail an annual plenary meeting alongside two
thematic working groups: one on the spectrum of security issues
encountered in cyberspace; and the other on capacity building programs
to help overcome the “digital divide” between developed and developing
states. Provisions would be made for input from non-governmental
stakeholders in civil society and the private sector given their critical role
in cyberspace activity. The modalities for such participation by NGOs were
a source of dispute throughout the group’s existence and the results,
although far from ideal from a transparency perspective, do ensure a
greater role for these non-governmental entities than had previously been
the case.
The blueprint for the “Global Mechanism” will still have to receive final
endorsement at an organizational meeting slated for March 2026, but the
prospects seem good that finally the UN will have a permanent
institutional home for consideration of cyber security issues.
The contrast between the norms of responsible state conduct and the
incidents of offensive cyber operations that violate them points to a basic
vulnerability of such accords. There is no real mechanism for the
enforcement of these norms, and as we have seen in other areas, the UN
Charter has a structural weakness in the veto rights granted to the five
permanent members of the Security Council.
These vetoes have protected permanent members from any action against their own
culpability for violations of international law. The creation of a permanent
UN body to address cyber security issues still suffers from an
“accountability deficit” in its lack of a specific measure for holding states
to account for their actions in cyberspace.
The NGO ICT4Peace, for which I am a Senior Advisor, proposed in 2020
the establishment of a “Cyber Peer Review” mechanism. Such a
mechanism would be modeled on the Universal Periodic Review
mechanism of the Human Rights Council. It would ensure a process
whereby states could regularly be assessed on the degree of their
compliance with the measures that they have agreed to. In the absence
of a process for holding states to account, it is probable that some states
will persist in conducting cyber operations in violation of the agreed
normative framework of restraint. The future meetings of the “Global
Mechanism” could be used to call out problematic behaviour by states.
Such challenges in the opaque realm of offensive cyber operations may
be difficult to sustain and would be vulnerable to being dismissed as just
disinformation by adversaries. It would be preferable if a form of
systematic peer review could be instituted and provide an incentive for
states to abide by the norms of responsible behaviour they have already
agreed to.”
ICT4Peace Foundation
Geneva, 27 October 2025
This blog post was published part of the “Canadian Security Interests and
Trump 2.0” Conference