Realizing NATO’s Women, Peace and Security Commitment in Practice | By Katharine A.M. Wright & Diana Morais

NATO is committed to integrating Women, Peace, and Security
(WPS) across all its tasks in the 2022 Strategic Concept and the
revised NATO Policy on WPS endorsed by Heads of State at the
2024 NATO Summit in Washington, D.C. The NATO Strategic
Concept referenced the Women, Peace and Security (WPS)
agenda for the first time in 2022. This was a significant move
since it elevated a gender perspective from the margins to a more
central position in the Alliance’s agenda. The revised NATO
Policy on WPS endorsed by Heads of State in 2024 reinforces
this. The endorsement reflects the reality that the WPS agenda,
and a gender perspective, is all the more relevant as the Alliance
returns to a primary focus on deterrence and defense following
Russia’s full-scale invasion of Ukraine in February 2022.1
Russia
is indeed waging a deeply gendered war in terms of the use of
conflict-related sexual violence, but also in its broader attack on
NATO values, including gender equality, through the promotion
of “traditional” family values.
This Policy Brief looks ahead to consider the question of what
NATO’s commitment to WPS, as outlined in the Strategic
Concept and the revised NATO WPS Policy, would look like if
it were realized in practice, specifically through the NATO WPS
Action Plan due for renewal in 2025. To do so, it engages with
how WPS and a gender perspective have been institutionalized
in NATO, noting that advances in this agenda have often not
taken the “usual” or proscribed route within the Alliance.
It is, therefore, an area ripe for innovation in terms of its
implementation. The WPS agenda is now in a strong position
at NATO with buy-in from its Member States.2
The next step
is realizing the strategic vision for WPS’ implementation
set out in the revised NATO WPS Policy (2024) through the
development of the NATO Action Plan on WPS due in 2025.
Such an approach would drive the Alliance’s long-term WPS
priorities into the future and ensure WPS’ institutionalization
at all levels, political and, especially, military. It must rely on
gender-responsive leadership to strengthen existing gender
expertise across the Alliance and ensure accountability.
Realizing Women, Peace, and Security as
an Asset for Operational Effectiveness
The WPS agenda emerged from UN Security Council Resolution
(UNSCR) 1325, adopted in 2000, and the nine follow-up
resolutions. It recognizes women’s agency and calls for women’s
participation in peace and security while acknowledging the
gendered impact of conflict, which often disproportionately
impacts women. These two aspects of the WPS agenda are
mutually reinforcing.3
The realization of UNSCR 1325 relied on
insiders and outsiders to the Security Council, including civil
society, which remain crucial knowledge brokers of WPS.4
NATO’s initial formal policy engagement with WPS in 2007
did not come with civil society involvement; rather, it built on a
concern related to the status of women in NATO forces on the
one hand and operational requirements vis-à-vis Afghanistan,
specifically counterinsurgency, on the other. This has set NATO
apart from the UN Security Council and other national and
regional level engagements with WPS that have been premised
on civil society consultation. It also demonstrates the nexus between WPS and a gender perspective in NATO’s approach.5
However, since 2014 the Alliance has consulted civil society
formally on its WPS policy, the first time it has done so on any
policy area.6
The Civil Society Advisory Panel (CSAP) on WPS
provides an important pool of expertise for NATO to draw on in
strengthening its approach to WPS.7
NATO’s engagement also did not come out of the blue. It
mapped onto decades of organizing within the Alliance to
strengthen the status of women in the armed forces of the
Alliance, including through the then Committee on Women in
NATO Forces (CWINF), which first met in 1961. It was through
CWINF that senior military women organized meetings in the
1960s on issues concerning the recruitment and retention of
women in the armed forces and successfully advocated for the
committee’s formal recognition in 1976. From 2002, CWINF led
on NATO’s WPS work, introducing the topic within the Alliance
and including the monitoring of the agenda’s implementation
among Allies. In 2009, CWINF became the NATO Committee
on Gender Perspectives (NCGP), recognizing its expanded
remit to monitor and support NATO’s implementation of WPS
by promoting the integration of a gender perspective into the
design, implementation, monitoring, and evaluation of policies,
programs, and military operations.
NATO’s engagement with WPS has relied upon decades of
military women’s advocacy within NATO and now benefits from
broader institutional buy-in, including from men. However, the
agenda has always gained traction when it has been perceived as
providing “added value” for the Alliance, including improving
operational effectiveness. For example, the Alliance would not
have engaged formally with the agenda when it did without its
involvement in Afghanistan through the International Security
Assistance Force (ISAF) from 2001-2014.8
Afghanistan saw
a “new” way of fighting with counterinsurgency operations
(COIN), given the reality that at the operational and tactical
levels there was a need to incorporate a gender perspective, as
the central focus of the campaign was the Afghan population.
This led to the deployment of Female Engagement Teams and
Cultural Support Teams, which instrumentalized gender as
a “new strategic asset,” demonstrating that the U.S. (and, by
extension, NATO) was “no longer fighting its battles with
outmoded methods.”9
In NATO’s current operations in Iraq,
Kosovo, and the Baltics, the military continues to have a key
role in implementing WPS, with states such as Canada driven
to lead on the agenda through Canada’s commitment to its own
Feminist Foreign Policy and because of the perceived added
value it provides to supporting operational effectiveness.10 This
is not to say there are no significant challenges to living up to
WPS in practice. For example, Canada was recently criticized for
not apologizing to an employee who had been sexually assaulted
by a NATO soldier while deployed as part of NATO’s Enhanced
Forward Presence in Latvia.11
WPS has also been seen as valuable to NATO in building
partnerships. A key example is NATO’s partnership with Sweden
before its bid for NATO membership. Sweden’s involvement
in ISAF helped shape NATO’s early views on WPS and gender
perspectives as operational assets.12 Despite Sweden’s nonalignment stance at the time, it deepened its partnership with
NATO through WPS, notably supporting gender integration in
Afghanistan and providing diplomatic backing for NATO’s WPS
efforts, such as at the 2012 Chicago Summit.13
However, Sweden’s case also reveals WPS’ vulnerability
within NATO. While WPS was central to the NATO-Sweden
partnership, any mention of it was omitted when Sweden began
formal NATO membership talks. This suggests that Sweden’s
Feminist Foreign Policy—an approach it pioneered and which
became a core aspect of its international stance, conflicted with
its NATO aspirations.14
Beyond Sweden, NATO partnerships have played an integral
role in the Alliance’s understanding of WPS, with initial NATO
WPS policies adopted jointly by the Euro-Atlantic Partnership
Council (EAPC).15 The agenda has thus provided “added value”
for NATO by providing a “safe” non-contentious issue and
diplomatic tool for NATO to build relationships with partners
whose interests might usually differ. For example, Austria proved
critical in supporting NATO’s adoption of a policy on WPS in

Japan, set geopolitically apart from NATO but increasingly
important for a potential pivot to the Indo-Pacific, initially built
its partnership with NATO on WPS, providing a Voluntary
National Contribution to NATO HQ to support NATO’s WPS
work in 2015 for two years.
Another area in which WPS has run counter to “business
as usual” at NATO, though again has added value, is with
respect to the role of civil society and promoting the Alliance’s
image in global politics. At a national level, and indeed at the
European Union level, civil society as knowledge brokers of
WPS have been heavily involved in developing WPS policies.
As a defense Alliance, such policy consultation requires a level
of transparency and openness that NATO did not engage in, in any policy area.16 Likewise, many civil society advocates of
WPS with roots in pacifism remained skeptical of NATO’s
engagement with WPS; consequently, there was no civil society
lobbying for NATO to engage with WPS at the time. Therefore,
NATO remained one of the exceptions among actors—states
and international organizations, implementing WPS only in
2014 and establishing a mechanism for formally consulting civil
society in the WPS policy-making process for the first time.
This has not been without challenges, but it represented a
significant commitment to open NATO’s WPS work to outside
scrutiny and a recognition that such consultation can strengthen
NATO’s policy.17
Institutional Drivers and Leadership on
WPS at NATO
On the political side, the creation of the Secretary General’s
Special Representative (SGSR) on the WPS position in 2012
has provided important impetus for NATO’s WPS work and has
been recognized by the UN as an example of best practices given
the high-level reporting of the role.18 Yet, the establishment of
the position was far from a done deal prior to the 2012 Chicago
Summit, and it took some NATO officials by surprise.19 Norway
offered to fill in the role from 2012-2014. At this time, institutional
challenges remained to fulfilling the SGSR’s mandate, including
the position within the NATO structure and rank. This was a
topic for discussion from 2014 to 2017, with the Netherlands
putting forward and sponsoring an ambassador for the position.
The high-level SGSR has successfully raised awareness of NATO’s
WPS work externally and taken the lead on WPS development
internally.20
The SGSR is now recruited via the regular NATO recruitment
process, which has meant bringing the funding for it “in-house”
from the civil budget. As part of the Office of the Secretary
General, this reflects progress on the WPS agenda at NATO.
The open competition also brings the process in line with the
recruitment for Assistant Secretary General positions, even if
certain Allies still dominate particular portfolios here, bringing
into question how “open” such competition is.21 In practice, most
Allies with a WPS National Action Plan have seen the benefit of
promoting the agenda as a foreign policy tool to support their
wider influence, including at NATO, so it remains a politically
lucrative position for Allies to secure.22
The SGSR on WPS remains the main point of accountability for
implementing WPS at NATO, as outlined in the current Action
Plan. However, the WPS agenda also spans across the areas of
responsibility of several Assistant Secretary Generals (ASGs),
from Innovation, Hybrid and Cyber to Defence Policy and
Planning and Operations and Public Diplomacy. Various WPSrelated initiatives have emerged from these ASGs, most recently,
a Public Diplomacy campaign to counter gender disinformation
on social media.23 That withstanding, to fully achieve the genderresponsive leadership committed to in NATO’s WPS Policy, and
to prevent the agenda from being siloed on the political side
or solely managed by the SGSR and their office, there is a need
for each ASG to have specific actions in the upcoming Action
Plan. In this way, each ASG will share accountability for the
implementation of the WPS agenda, even if the SGSR remains
the driver and a crucial guide for coherent WPS implementation
at NATO.
On the military side, the NATO Committee on Gender
Perspectives (NCGP) serves as a prime example of the critical
role institutional drivers play in successfully implementing the
WPS agenda within NATO. As an advisory body to the Military
Committee (MC), the NCGP has been at the forefront of
institutionalizing gender perspectives and operationalizing the
NATO WPS policies.24
It began back in the 1960s, exploring every avenue through
which women could make the fullest possible contribution to
NATO by serving in their national armed forces. Its efforts led to
the establishment of the International Military Staff (IMS) Office
of the Gender Advisor (GENAD) in 1998. Its role has expanded
in recent years and now also supports the implementation of
WPS and the integration of a gender perspective within the
IMS, marking a significant milestone in institutionalizing
gender perspectives.25 Since 2005, it has actively championed
the creation of a Military Committee policy document on
gender perspectives within NATO, a long-fought goal that was
realized in the adoption of the NATO WPS policy last year. This
achievement underscores decades of persistent advocacy to fully
integrate a gender perspective into NATO’s activities, missions,
and operations.
However, for the military, there are additional practical challenges
in implementing WPS and incorporating a gender perspective,
with the primary issue being inadequate resourcing. Gender
Advisors (GENADs) positioned throughout NATO, including
within the International Military Staff and Strategic Commands, are understaffed and, therefore, unable to fully meet the political
ambitions of the WPS agenda.
Putting WPS Into Practice Through the
Revised NATO Action Plan
WPS remained on the margins of NATO’s core business,
although, as demonstrated here, it has provided significant value
to the Alliance in several areas. Its elevation to a deliverable
of the Strategic Concept and the revised NATO WPS Policy
providing a strategic vision to the agenda indicates a novel phase
in NATO’s engagement and an attempt to make WPS part of the
Alliance’s core business.
The Strategic Concept states that NATO “will promote good
governance and integrate climate change, human security, and
the Women, Peace, and Security agenda across all our tasks. We
will continue to advance gender equality as a reflection of our
values.”26 While the inclusion of WPS is significant though not
guaranteed, how it has been framed in the Strategic Concept
affects how WPS will be understood within NATO moving
forward. So far, the implementation of the WPS agenda and a
gender perspective have added value to NATO, particularly by
improving operational effectiveness and supporting women in
NATO forces.27 As the Secretary General has stated, integrating a
gender perspective is “not only the right thing to do but the smart
thing to do.”28 The absence of WPS in the Strategic Concept’s
sections on Deterrence and Defence, as well as Crisis Prevention
and Management, is significant since the document guides the
entire Alliance. The updated NATO WPS Policy addresses this
gap by aligning it with NATO’s core tasks and the four pillars
of the WPS agenda. However, the Strategic Concept missed the
opportunity to include this directly, meaning those not fully
convinced of WPS’s importance would need to consult the WPS
Policy separately to understand its relevance.
Crucially, the NATO Policy on WPS introduces the concept
of gender-responsive leadership “to ensure NATO leaders
strengthen their gender expertise, work towards gender equality
and are accountable for the implementation of the WPS
Agenda.”29 Such an approach will be key to realizing an effective
WPS agenda at NATO and should be integrated into the Action
Plan. It should start from the top down, with the Secretary
General mainstreaming WPS into public-facing remarks as a
matter of course, not as an exception. In addition, and as outlined
earlier, clear lines of reporting on WPS should be identified for
all the Assistant Secretary Generals’ portfolios. This will help
address the weaknesses in WPS in the Strategic Concept and
draw attention to the importance of NATO’s revised WPS Policy,
which is by far the most comprehensive policy to date and sets an
ambitious agenda for the Alliance.
At present, there is value in ensuring the integration of a gender
perspective is reflective of NATO’s WPS priorities and moving
beyond the institutional siloing of this topic to ensure the whole
of NATO is committed to its implementation across political
and military structures. Establishing the SGSR WPS position
has been fundamental to moving this agenda forward, as well as
approving the latest NATO WPS Policy, which has provided an
overarching strategic vision. But now it needs to be translated
into practice via the Action Plan, and a crucial part of this will be
the actions of NATO leadership across the political and military
structures.
To summarize, advancements in the integration of a gender
perspective in NATO are tempered by remaining challenges. The
inclusion of WPS in the Strategic Concept is highly significant.
However, while it is referenced in relation to all of NATO’s tasks,
it is specifically mentioned only in the context of Cooperative
Security, with no direct mention in the sections of the other
core tasks (on Deterrence and Defence or Crisis Prevention and
Management). This deficiency should be addressed. Secondly,
as the revised NATO Policy on WPS (2024) draws attention to,
translating the policy commitment of WPS into practice must
emphasize WPS’ added value to NATO, including promoting
gender-responsive leadership by establishing clear accountability
for the Assistant Secretary Generals and their portfolios in
mainstreaming WPS across all NATO activities. Finally, it is
essential to draw on the expertise of the Civil Society Advisory
Panel (CSAP) on WPS and the NATO Committee on Gender
Perspectives (NCGP).
Futureproofing WPS at NATO
As NATO’s engagement with WPS to date has demonstrated, this
is a policy area ripe for innovation that adds considerable value
to the Alliance at all levels. The next step in supporting NATO’s
WPS commitments is to translate Strategic Concept statements
and NATO WPS policy commitments into practice through the
development of an effective WPS Action Plan. In this regard, four steps would contribute towards NATO’s
efforts to put WPS into practice in developing the new NATO
WPS Action Plan, which would center on the added value of
WPS for the Alliance.

The Secretary General and NATO’s senior leadership should
integrate WPS into their public-facing remarks as a matter of
routine, given the demonstrated applicability of the agenda to
all of NATO’s tasks.

Align efforts with the strategic vision that prioritizes
gender-responsive leadership and ensures accountability in
implementing WPS across NATO’s political and military
structures.

NATO’s political and military structures should better leverage
the expertise of the NATO Civil Society Advisory Panel on
WPS and NCGP, respectively, using it as a resource to support
gender-responsive leadership in their work.

Ensure that policy commitments are translated into concrete
actions by providing the NATO Military Authorities with
adequate resources—especially in terms of human resources
and dedicated gender structures—to implement these
initiatives effectively.

By Kulani Abendroth-Dias and Carolin Kiefer

If World War III will be over in seconds, as one side takes control of the other’s systems, we’d better have

Data, the food of all algorithms, lie at the core of cohesive EU and NATO AI strategies. Such strategies must encompass the regulation of data in high- and low-risk technologies with

and Romania have tested and often deployed AI and ML facial recognition tools, many of which were developed in the United States and China, for predictive policing and border control.3 AI and ML systems aid in contact tracing and knowledge sharing to contain the COVID-19 virus.4 However, the civilian and military strategies that drive use of AI and ML for the collection and use of data diverge across the member states of the European Union and the North      a greater understanding of how data feed AI and ML technologies and systems, the results they produce become skewed. For example, a facial analysis and recognition system insufficiently trained to analyze and recognize women or people of color will often misidentify people in these populations, which could lead to inaccurate criminal profiling and arrests.7 Machines don’t make errors, but humans do. Policymakers need to rapidly identify parameters and systems of governance for these technologies that

maximize their efficiency while protecting civilian rights.

Growth in the development of AI-driven technologies has been exponential, but strategies to regulate their implementation have yet to catch up. The European Union and NATO need to develop coordinated, comprehensive, and forward-looking strategies based on data protection protocols to regulate AI use and deployment to counter myriad threats. Such strategies will be critically important if the transatlantic alliance is to adapt a common defense system to evolving threats in the digital age.    Beyond Definitions

AI and ML are changing the security landscape-for example, by the deployment of disinformation to undermine political participation or of unmanned autonomous vehicles (UAVs), which may or may not operate as lethal autonomous weapons systems (LAWS). The states that are party to the Group of

Governmental Experts (GGE) on Lethal Autonomous Weapons

the smarter, faster, more resilient network.       dual uses. They should guide policies governing predictive

or delivery within the European Union, Amazon    policing, border surveillance, facial analysisand countering disinformation.6     and recognition now sells facial recognition cameras for door

locks, webcams, home security systems, and office        To regulate data use effectively, policymakers need to attendance driven by artificial intelligence (AI)        better understand the technical, political, economic and

and machine learning (ML)-powerful tools with civilian 2  social risks and biases in data collection methods. Without and military purposes. Germany, France, Spain, Denmark

Atlantic Treaty Organization (NATO).5

(LAWS), which aligns its work with the Convention on Certain Conventional Weapons (CCW), have devoted considerable attention to defining autonomous weapons. Unfortunately, the group has not yet paid enough attention to the data. Prolonged focus on what constitutes LAWS rather than the data that drive them impedes the important investigation of how best to regulate the technologies’ rapid development and use for security and defense. Discussion of the types, limits, and biases of data that drive AI and ML is pertinent throughout the myriad sectors in which they find application.8

Recently, the GGE took steps to move the debate from definitions of autonomous systems to why data matter. In 2020, it decided that the 11 guiding principles that frame the development and use of LAWS needed no further expansion.9 The group agreed to give greater attention to how the principles can be unpacked. It decided to distinguish between high- and low-risk AI technologies and gain a better understanding of dual-use technologies.10 Differentiating between uses for civilian and military operations should focus on how data will be mined and drive algorithms at both levels.11 NATO and the European Union should lead in facilitating these discussions and regulations. 

Data Governance

According to the European Commission’s February 2020 white paper on artificial intelligence, “Europe’s current and future sustainable economic growth and societal well-being increasingly draws on value created by data…. AI is one of the most important applications of the data economy.”12 However, the report concludes, for AI to “work for people and be a force for good in society” it must be trustworthy.13 It highlights “trustworthy AI” 27 times in its 26 pages.

Governance of data is key to this trust.14 The EU General Data Protection Regulation (GDPR) was a step in the right direction, but it needs to be expanded to cover AI and ML data collection and use in national and international security contexts. Close consultation and data coordination between the European Union and NATO is integral in this regard.

An understanding of who drives the development of AIdriven technologies for European security and how they are funded can illuminate the political, technical, and social, and legal bottlenecks confronting EU and NATO data regulation, both in the member states and at a supranational level. While the defense sector has traditionally driven technology innovation, private companies have taken the lead in recent years. 15 According to the OECD, Google, Microsoft, Amazon, and Intel have spent more than $50 billion a year on digital innovation.16 This sum dwarfs the ‚Ǩ13 billion budgeted by the European Defense Fund (EDF) for 2021-27 – for defense spending in general, not solely for AI-driven technologies.17 NATO and the European Union should pay particular attention to these private-sector actors when developing policies for data protection and strategies to encourage US and European technological innovation. NATO and the European Union should work with the CCW GGE to determine clear operational distinctions between the commercial and military uses of data for AIdriven technologies.18 NATO and the European Union need comprehensive, legally enforceable AI strategies to regulate the use of data and the integrity of information networks to better protect their citizens while keeping the Alliance agile.

The Way Forward

In EU and NATO contexts, the development and implementation of dual-use technologies and cyberprotection policies remain fragmented. This fragmentation could undermine the ability to respond to evolving threats to European security and stability. Examples abound: Cambridge Analytica’s involvement in Britain’s Leave Campaign, radicalization via social media, the politicized use of data via hybrid-use platforms to influence behavior (from political participation to violent action), and targeted cyber-attacks and disinformation campaigns in the Visegrad Four and the Baltic states.19 Therefore, coherent EU and NATO AI strategies require the regulation of the data that drive emerging technologies. Regulation to promote network integrity and protect data access must be key tenets of EU and NATO strategies to deploy AI that can react faster and more effectively in the face of new security threats.

AI and ML systems are valuable, as demonstrated by their use in contact tracing and knowledge sharing in the search for a cure during the Covid-19 pandemic.20 For the transatlantic relationship to thrive, NATO and the EU must work together to develop coordinated AI strategies that address appropriate use and misuse of data. As the EU and NATO develop these strategies, they should focus on five activities:

Govern the use of data in dual-use technologies.

While AI strategies may sound exciting and innovative to policymakers and the general public, responsible data use sounds less so. Yet it is essential. EU and NATO strategies need to distinguish between high- and low-risk technologies, dual- and hybrid-use platforms, and the types, limits, and mediums by which data can be collected and anonymized (or at least kept confidential) for civil and military uses. These limits need to be developed and regulated in discussions with civilian and military actors who are mining data across sectors, from the traditional security and military arena to healthcare, logistics, and entertainment companies. Discussions should include how the rights of citizens and those residing in NATO and EU countries-e.g., lawful migrants, asylum seekers, refugees-will be protected.

Acknowledge bias in datasets.

There should be a comprehensive discussion on how bias in datasets influences the training of algorithms, which in turn influences security targets and undermines the integrity of a system. Policymakers, human rights actors, and technology developers should be in the room for this discussion. An awareness of these biases within security forces can help them better evaluate the outcomes the algorithms produce, interpret targets with caution, avoid errors, and generate more effective responses.

Ensure purpose-limited data collection and sharing.

Personal data collected and tracked for specific purposes (e.g., contact tracing during a pandemic) should generally not be shared and used for other purposes. Where an overlap in data collection is deemed necessary for EU-NATO security purposes, tight regulations for civilian protection should spell out where, with whom, and for how long the data can be stored, with strong legal and operational deterrents for backdoor access to data. Private-sector companies should limit how data are used to influence behavior: Should they be used in political campaigns the same way that they are used to nudge consumer behaviors on what to buy? The European Union’s GDPR sets up important rules in this regard. It can be viewed as the cornerstone of an EU-NATO strategy for the development and regulation of AI for security and defense.

Adapt traditional defense and deterrence strategies to the digital age.

The evolving nature of security threats in the digital age calls into question traditional strategies of defense and deterrence. Collaboration between NATO, the European Commission, the European Defence Agency (EDA), the Permanent Structured Cooperation (PESCO), the Coordinated Annual Review on Defense (CARD) and technology developers should focus on efficiency-trimming current weapons systems and technologies used by the European Union on the battlefield and in the cyber realm while using AI and ML to inform strategy. The weaponized use of social media data must be addressed, not solely via counternarratives but by working in concert with social media companies to develop AI and ML techniques to identify and shut down fake news at the source. The integrity of networks set up by actors outside of NATO member states needs to be raised as a security concern as well, including incentives to drive the local business development of such networks.

Build trust via counter-AI agencies to protect citizen rights and detect AI-driven forgeries.

Agencies that currently promote the responsible use of AI need to work in tandem with NATO and EU agencies to develop comprehensive AI strategies. The strategies should promote digital literacy, advance critical thinking through online modules, and publicize the precautions NATO and the European Union are taking to protect citizen data in order to build public trust. Partnerships between EU, NATO, and such agencies need to go beyond traditional NGO-security agency relationships to integrate AI protection mechanisms into security policy itself. Ideally, these organizations would work with NATO partner countries to better identify targets, weaknesses, and priorities to build resilient intelligence architectures.

Map the development and use of AI-driven technologies across EU and NATO member states.

NATO security operations are in place at member state borders. However, most of the AI technologies being developed, test, or adapted are deployed within France and Germany, key EU member states. AI-driven security threats differ across states, especially disinformation. For example, the content, medium, and speaker of disinformation shared in the Czech Republic may differ considerably from disinformation shared in Germany. Adapting traditional deterrence strategies to the digital age requires an understanding of the contextbased nature of these threats. It is therefore integral to include experts across the EU and NATO member states in the development and implementation of AI strategies. A comprehensive mapping of the security threats faced-and development and use of AI-driven technologies to combat such threats across EU and NATO member states-can help better train personnel and develop more targeted solutions and localized data protection policies.

Conclusion

The digital industry is already transforming the Alliance. NATO is essential to setting up a coordinated structure to develop and regulate AI- and ML-driven technologies for NATO members’ security and defense. While sociopolitical and economic priorities in the development and regulation of AI vary across sectors and countries, awareness of the use and misuse of data in driving AI-and ML-driven technologies is a common thread that binds these debates together. The use of data fed into a system run by AI and ML technology can have vast implications for the nature of future security threats and the development of technologies to combat them. Cohesive EU and NATO strategies for AI will determine how strong and agile the Alliance will become.

References

Pedro Domingos, The Master Algorithm: How the Quest for the Ultimate Learning Machine Will Remake Our World (New York: Basic Books, 2015).

See Daniel S. Hoadley and Nathan J. Lucas, Artificial Intelligence and National Security (Washington DC.: Congressional Research Service,

2018); Greg Allen and Taniel Chan, Artificial Intelligence and National Security (Cambridge, MA: Belfer Center for Science and International Affairs, 2017). Artificial intelligence comprises a vast number of fields, including machine learning, natural language processing, robotics, computer vision, and knowledge representation and reasoning. In this policy brief, the authors largely refer to the use of AI- and ML- driven technologies for EU and NATO security and defense.

Steven Feldstein, The Global Expansion of AI Surveillance (Washington, DC: Carnegie Endowment, September 2019).

The Cybersecurity Strategy of the Visegrad Group Countries Coronavirus, Artificial Intelligence web page (April 12, 2020).

Raluca Csernatoni, An Ambitious Agenda or Big Words? Developing a

European Approach to AI, Egmont Security Policy Brief No. 117 (Brussels: Egmont Royal Institute for International Relations, November 2019).

Michael Chui et al., Notes from the AI Frontier: Insights from Hundreds of Use Cases (New York: McKinsey Global Institute, 2018).

Philipp Gr√ºll, “Germany’s Plans for Automatic Facial Recognition Meet Fierce Criticism,” EURACTIV (January 10, 2020).

Ajay Agrawal, Joshua Gans and Avi Goldfarb, Prediction Machines: The Simple Economics of Artificial Intelligence (Cambridge, MA: Harvard Business Press, 2018).

See the UN’s 1980 Convention on Certain Conventional Weapons.

German Federal Foreign Office, Chair’s Summary: Berlin Forum for Supporting the 2020 Group of Governmental Experts on Lethal Autonomous Weapons Systems (Berlin: German Federal Foreign Office, April 2020).

United Nations Institute for Disarmament Research, The Human Element in Decisions About the Use of Force (Geneva: UNIDIR, March 2020).

European Commission, On Artificial Intelligence: A European

Approach to Excellence and Trust, white paper, COM (2020) 65 final (Brussels: European Commission, February 19, 2020), p. 1. See also European Commission, A European Strategy for Data, COM (2020) 66 final (Brussels: European Commission, February 19, 2020).

EC, On Artificial Intelligence, p. 25.

Ibid.

Dieter Ernst, Competing in Artificial Intelligence Chips: China’s Challenge amid Technology War, Special Report (Center for International Governance Innovation, March 26, 2020).

Organization for Economic Cooperation and Development, Private Equity Investment in Artificial Intelligence (Paris: OECD, December 2018).

European Commission, European Defence Fund (Brussels: European Commission: March 20, 2019). Arguably, Washington would do well not to view the EDF with suspicion and skepticism but rather as a vehicle to stimulate more transatlantic discussion on “home-grown” innovation and development.

Daniele Amoroso et al., “Autonomy in Weapon Systems: The Mili-tary Application of Artificial Intelligence as a Litmus Test for Germany’s New Foreign and Security Policy,” Democracy Vol. 49 (Heinrich B√∂ll Foundation, 2018).

Marek G√≥rka, “The Cybersecurity Strategy of the Visegrad Group

Countries,” Politics in Central Europe Vol .14, No. 2 (2018), pp. 75-98. See also Alistair Knott, “Uses and Abuses of AI in Election Campaigns,” presentation, N.d.

Council of Europe, AI and Control of Covid-19.