« AI and European Union: priority to ethics in the context of use of AI as a geopolitical tool «
« AI and European Union: priority to ethics in the context of use of AI as a geopolitical tool
Contribution made by Dr Pierre-Emmanuel Thomann, Eurocontinent, to the
VI International Conference “Tangible and Intangible Impact of Information and Communication in the Digital Age” to be held in Khanty-Mansiysk, Russian Federation, on June 18-20, 2024, within the framework of the UNESCO Intergovernmental Information for All Programme and the XV International IT Forum.
The conference was convened by the Russian UNESCO IFAP Committee and the Interregional Library Cooperation Centre in cooperation with the Government of the Khanty-Mansi Autonomous Area – Ugra, the Permanent Delegation of the Russian Federation to UNESCO, and the Commission of the Russian Federation for UNESCO.
Introduction
This contribution aims to make a provisional diagnosis on the capacity of EU, claiming to give priority to ethics regarding the rapid AI developments, to fight MUAI (Malicious Use of Artificial Intelligence) in the context of use of AI as a geopolitical tool. It is necessary to assist in reducing the gap between the pace of development in the field of AI and related technologies and the ability of the society to comprehend the changing reality, and to propose recommendations to contribute to the accomplishment of international and national policies in the field of building the information society and knowledge societies for better global geopolitical balance and stability and better human control of its destiny.
Malicious use of Artificial intelligence (MUAI) can be used as a geopolitical weapon to destabilize the system of international relations to modify the geopolitical hierarchy and implement new imperialist agenda and projects, with the use of tools in order to achieve the control of populations through cognitive warfare, « digital colonization », global monopoly or hegemony on IA technology and clouds (fuel of AI), extraterritorial laws on data (digital imperialism), AI engineered programs to control territory and battlefield (Boulanger 2020, 2016), with the development of AI engineered geospatial intelligence (GEOINT ).
The AI Act of the European Union is the first-ever comprehensive legal framework on AI worldwide. The aim of the new rules is to foster trustworthy AI in Europe and beyond, by ensuring that AI systems respect fundamental rights, safety, and ethical principles and by addressing risks of very powerful and impactful AI models.
The main focus of the European Union regarding AI and digitalization was so far on its ethical, normative and economic aspects in the context of regulating the EU common market. However, no systemic analysis has been conducted of the ability of the EU to position itself regarding the geopolitical challenges posed by digitalization and AI, and particularly concerning MUAI and the threats to IPS in the context of greater power rivalries and the diversity of views of EU member states towards the strategic aims of the European project.
EU AI ACT : overview of EU legislation regarding risks associated with AI
The EU’s 27 member states have unanimously endorsed the AI Act, affirming the political agreement reached in December 2023. Similarly, the Internal Market and Civil Liberties Committees in the European Parliament voted 71-8 (7 abstentions) to approve the result of negotiations with the member states on the AI Act.
The EU official documents describe the new EU legislation as followed: “ The AI Act is the first-ever comprehensive legal framework on AI worldwide. The aim of the new rules is to foster trustworthy AI in Europe and beyond, by ensuring that AI systems respect fundamental rights, safety, and ethical principles and by addressing risks of very powerful and impactful AI models “[i]
What is EU posture regarding the risks associated with AI in this new legislation ?
The risks are divided into different categories: unacceptable risks, high risks; limited risks and minimal risks
Unacceptable risks of AI according to EU (Prohibited AI systems (Chapter II, Art. 5)
AI systems:
-deploying subliminal, manipulative, or deceptive techniques to distort behaviour and impair informed decision-making, causing significant harm.
-exploiting vulnerabilities related to age, disability, or socio-economic circumstances to distort behaviour, causing significant harm.
-social scoring, i.e., evaluating or classifying individuals or groups based on social behaviour or personal traits, causing detrimental or unfavourable treatment of those people.
-assessing the risk of an individual committing criminal offenses solely based on profiling or personality traits, except when used to augment human assessments based on objective, verifiable facts directly linked to criminal activity.
-compiling facial recognition databases by untargeted scraping of facial images from the internet or CCTV footage.
-inferring emotions in workplaces or educational institutions, except for medical or safety reasons.
-biometric categorisation systems inferring sensitive attributes (race, political opinions, trade union membership, religious or philosophical beliefs, sex life, or sexual orientation), except labelling or filtering of lawfully acquired biometric datasets or when law enforcement categorises biometric data.
-‘real-time’ remote biometric identification (RBI) in publicly accessible spaces for law enforcement, except when:
- targeted searching for missing persons, abduction victims, and people who have been human trafficked or sexually exploited;
- preventing specific, substantial and imminent threat to life or physical safety, or foreseeable terrorist attack; or
- identifying suspects in serious crimes (e.g., murder, rape, armed robbery, narcotic and illegal weapons trafficking, organised crime, and environmental crime, etc.).
-Using AI-enabled real-time RBI is only allowed when not using the tool would cause harm, particularly regarding the seriousness, probability and scale of such harm, and must account for affected persons’ rights and freedoms.
-Before deployment, police must complete a fundamental rights impact assessment and register the system in the EU database, though, in duly justified cases of urgency, deployment can commence without registration, provided that it is registered later without undue delay.
-Before deployment, they also must obtain authorisation from a judicial authority or independent administrative authority[ii], though, in duly justified cases of urgency, deployment can commence without authorisation, provided that authorisation is requested within 24 hours. If authorisation is rejected, deployment must cease immediately, deleting all data, results, and outputs.
High risks of AI according to EU
High-risk AI systems will be subject to strict obligations before they can be put on the market: adequate risk assessment and mitigation systems; The AI systems identified as high-risk by EU include AI technology used in:
-critical infrastructures (e.g. transport), that could put the life and health of citizens at risk;
-educational or vocational training, that may determine the access to education and professional course of someone’s life (e.g. scoring of exams);
-safety components of products (e.g. AI application in robot-assisted surgery);
-employment, management of workers and access to self-employment (e.g. CV-sorting software for recruitment procedures);
-essential private and public services (e.g. credit scoring denying citizens opportunity to obtain a loan);
-law enforcement that may interfere with people’s fundamental rights (e.g. evaluation of the reliability of evidence);
-migration, asylum and border control management (e.g. automated examination of visa applications);
-administration of justice and democratic processes (e.g. AI solutions to search for court rulings).
All remote biometric identification systems are considered high-risk and subject to strict requirements. The use of remote biometric identification in publicly accessible spaces for law enforcement purposes is, in principle, prohibited. Narrow exceptions are strictly defined and regulated, such as when necessary to search for a missing child, to prevent a specific and imminent terrorist threat or to detect, locate, identify or prosecute a perpetrator or suspect of a serious criminal offence. Those usages is subject to authorisation by a judicial or other independent body and to appropriate limits in time, geographic reach and the data bases searched
EU AI legislation and defence and security
The EU AI act establishes three main exemptions where the AI Act does not apply[iii]:
- AI systems used exclusively for national security purposes.
- AI systems placed on the market or put into service exclusively for military, defence, or national security purposes.
- AI systems not placed on the market or put into service within the EU but whose outputs are used exclusively for military, defence, or national security purposes.
Examples and Implications
1. General Exemption for National Security: A practical example includes an AI system developed by a member state for real-time threat assessment and surveillance. This system might analyze data from social media, satellite imagery, and other sources to identify potential security threats. The exemption ensures member states can leverage AI for national security without AI Act constraints.
2. Exemption for Military and Defence Purposes: Consider an AI-enhanced unmanned aerial vehicle (UAV) used for reconnaissance missions in conflict zones. This UAV, equipped with AI for real-time data analysis and decision-making, falls outside the AI Act’s scope when used exclusively for defence purposes. The exemption supports technological advancement in military capabilities but raises ethical considerations, particularly regarding autonomous decision-making in lethal force deployment.
3. AI Systems Not Marketed or Put into Service in the EU but Used for Military or National Security: An example is the deployment of an AI-based cybersecurity defence system, procured from a non-EU country, to protect national infrastructure against cyberattacks. This exemption acknowledges the global nature of AI development and the necessity for states to access and utilize cutting-edge technologies for security purposes.
We examined EU main features of its legislation to fight against the risk associated with AI, but is EU able to face the more acuter problem of use of AI as a geopolitical weapon in the new geopolitical configuration?
AI and geopolitical diagnosis, Strategic MUAI and international system
The world is faced with increasing geopolitical fragmentation, with the multiplication of actors, reinforcement of the power gaps between states, and changes in previous geopolitical hierarchies. Moreover, geopolitical confrontation is increasingly taking place in the theatre of hybrid warfare, including psychological warfare.
In the context of the rapid development of AI technologies in recent years, the possible use of AI as a geopolitical weapon through the destabilization of IPS (international psychological security) may contribute to determining the international order in the 21st century and the emergence of “digital empires”), thus accelerating the dynamics of the previous cycle in which technology and power have mutually reinforced each other. This will transform some of the paradigms of geopolitics through the emergence of new relationships between territories and their populations, spatio-temporal dimensions and immateriality
The systemic nature and effect of strategic MUAI (Malicious Use of Artificial Intelligence) would be made possible by an increase in the actors’ room for manoeuvre in space and time. Great powers that can implement AI-enhanced strategies leading to multiple areas of supremacy in spatial dimensions such as the land, maritime, air, cybernetic, spatial, and cognitive domains and time gains, with their anticipation capacity favoured by predictive analysis, could lead to an overthrow of the international order or make stabilization impossible. (Diagram 1)
We will describe this process as the entry into space-time geopolitics, which requires the mastery of space and time at levels unprecedented in history and which might escape any moderation exercised by a multilateral system. We are already seeing the beginnings of this era with the impotence of multilateral bodies such as the UN, the EU, NATO, and the OSCE in Europe.
AI as space–time weapon of conquest
For a state or an alliance of states that possess the whole range of tools of power augmented by the mastery of AI, and are able to act in all the spaces of confrontation (land, air, sea, space, cyberspace, human brain), there is a great risk of being tempted to engage in international expansionist policies, thanks to the use of AI as a space-time weapon of conquest.
This will lead to an increase in the fragmentation of the geopolitical world, very profound destabilization of international relations, and an increase in power inequalities
Strategic MUAI is likely to have a decisive effect on the evolution of the geopolitical configuration, leading to a reinforcement of hierarchies and inequalities and possibly a new America-China bipolarity. It may also lead to an era where the notion of relative stability no longer exists. Following the rise of strategic MUAI, conflicts may become permanent and the current multilateral bodies, already largely obsolete in terms of containing crises, will be bypassed by decisive state and private actors engaged in geopolitical rivalry. These actors will act without any restrictions or control by the people.
As Raymond Aron has pointed out, “all international order, up to the present day, has been essentially a territorial order” (1962).
There can be no efficient way of containing conflict through a normative international system without an underlying spatial and geopolitical order accepted by the great powers. In a configuration where great powers agree on their respective red lines and zones of influence, that is, a spatial order that is necessarily precarious but allows room for dialogue and cooperation in multilateral fora, it would be possible to stem the destructive potential of IPS strategic MUAI. Alternatively, a scenario in which the international system is destroyed or ossified into an extremely unequal hierarchical order that leads to violent and sometimes extreme opposition might be favoured
The cognitive warfare as category of MUAI
The emergence of the concept of « cognitive warfare » (CW) (François du Cluzel, 2020)[iv], adds a third combat dimension on a battlefield: cognitive dimension is added to the informational and physical dimensions. As a result, we have now different geopolitical spaces of competition like land, maritime, air, cybernetic, spatial and cognitive domains. The « Human » might be considered as the sixth domain of operations (the fiv other: air, land, sea, space and cyber). From this perspective, cognitive war, thanks to scientific progress in nanotechnology, biotechnology, information technology and cognitive science (NBIC), combined with Artificial Intelligence, Big Data, and the increasing dependence on digital space by populations (internet, social networks), is targeting the brain of humans within populations. Cognitive warfare goes beyond information warfare since its objective is not to influence what people think but the way they think. It has the potential to transform a whole nation into a colony from a third state, or to disrupt a whole nation and its territory. With the emergence of the cyberspace theatre with the use of internet, the confrontation of narratives and antagonist ideas between rival powers or terrorist groups though artificial intelligence (AI) tools to change the way people think is permanent with no time and space limit as it takes place at different global, regional, national, local scales.
Studies devoted to the geopolitical implications of strategic MUAI and its implications for the EU and international psychological security are currently lacking. Analysis of the risks of MUAI in international relations tends to be focused on threats to democracy and the use that non-democratic regimes can make of it. The link between AI and international relations and the possible consequences the former might have in systemic terms (i.e., the mutation of the geopolitical configuration), is awaiting investigation
EU and power politics in the context of use of AI as a geopolitical tool
The EU has acknowledged the return of power politics in a contested multipolar world. It has emphasized that it needs to prepare for fast-emerging challenges because its strategic competitors are actively undermining its secure access to the maritime, air, cyber, and space domains.
In the cyber domain, the EU wants to develop and make intensive use of new technologies—notably quantum computing, AI, and big data—to achieve comparative advantage (e.g., in terms of cyber-responsive operations and information superiority). It also needs to maintain its excellence in ensuring autonomous EU decision-making, including that based on geospatial data.
However, the EU has not changed its doctrinal position on multilateralism and still refuses to accept the multipolar model. It promotes strategic autonomy, but it considers itself complementary to NATO and the US as its main strategic partner. It is therefore aligned de facto with the unipolar objectives of the US and will continue to cooperate in areas like respective security and defence initiatives, disarmament and non-proliferation, the impact of emerging and disruptive technologies, climate change and defence, cyber defence, military mobility, countering hybrid threats including foreign information manipulation and interference, crisis management, and relationships with strategic competitors (principally, Russia and China).
The EU does not consider the systemic implications of strategic MUAI on the geopolitical configuration and how it might acquire greater autonomy, especially from the US (on which it is very dependent); nor does it advocate a more stable and balanced international system. Here again, the EU’s close alliance with the US risks geopolitical fragmentation into antagonistic blocs at a global level.
EU and sovereignty
In these EU legislative proposals, the question of preserving the sovereignty of EU member states does not include the establishment of a European preference on the acquisition of systems using AI, or restrictions on products that comply with the standards but come from third countries. There are also no proposals on how to address the issues of AI in relation to power rivalries and its impact on its external relations
EU in mainstrean geopolitical configuration against Russia and China
EU-US technological/geopolitical alliance
Faced with the potential emergence of a new bipolarity on AI, dominated by the US and China, the EU apart from the ethical issues with which it intends to differentiate itself, and more sovereignty on data, a non-alignment posture with those states that do not wish to be sucked into this limitless confrontation is not taken into consideration. The resilience of the EU is not addressed in terms of the balance of power. The EU thus favors a technological alliance with the US and consequently risks reinforcing the global geopolitical fragmentation into antagonistic blocs. Although there are still many disagreements between the EU and the US, this approach may be an obstacle to a global negotiation on effectively fighting against MUAI at the tactical level, and particularly at the strategic level.
Although it has been mentioned in a fragmented way, no systemic analysis has been conducted of the ability of the EU to position itself regarding the geopolitical challenges posed by Cognitive Warfare, digitalization and AI, and particularly concerning MUAI and the threats to IPS in the context of greater power rivalries and the diversity of views of EU member states towards the strategic aims of the European project. The EU has also stressed the danger of political polarization through MUAI (European Parliament, 2019)[v]; however, if the EU promotes an obsolete model of development and is engaged in exclusive alliances against the emergence of multipolarity, there is a risk that this might in itself promote geopolitical polarization.
The EU is embedded in the mainstream geopolitical configuration as a sub-element of the US and NATO geopolitical priorities regarding AI-engineered global informational and cognitive warfare (Chessen, 2017)[vi] against China and Russia. The European parliament has stressed that if “AI in the field of defence is essential for ensuring European strategic autonomy in capability and operational areas it recognises the role of NATO in promoting Euro-Atlantic security and calls for cooperation within NATO for the establishment of common standards and interoperability of AI systems in defence; stresses that the transatlantic relationship is important for the preservation of shared values and for countering future and emerging threats”(European Parliament, 2020)[vii]. (Diagram 2)
EU as a digital colony of US
EU is positioning itself as a neutral regulator (competition principle) on activities on its territory but not as a global and sovereign power
This approach cannot succeed in an open system where geopolitical competition prevails
This would work if all actors would respect same EU rules (free and fair competition, free trade, budgetary austerity, environmental and social standards…)
USA and China are locked in a geopolitical competition and do not respect these rules
Technological national champions and flagship companies cannot emerge in EU like in US and China (state support and subsidies, government and state orders, state support for market expansion, civilian-military synergy…)
EU has excluded regulation of AI systems in defence and security, and this means that EU cannot be a real geopolitical actor and with its complementary with NATO, EU will be prisoner of NATO use of AI as a geopolitical tool (like Cognitive war)
In the end EU is a digital colony of US and cannot claim to have a geopolitical role and have an independent approach in AI since
-GAFAM are all US based
-Clouds act (US companies abroad transfer data (on EU citizens and companies) on US authorities
-US Extraterritorial laws
-NSA backdoors
Geopolitical fragmentation: obstacle to cooperation in MUAI
However, this is unlikely to contribute to the building of common global alliances to counter the threat of MUAI to IPS. This may lead to total and permanent war and place humanity in danger.
Following the rise of strategic MUAI and Cognitive Warfare, conflicts may become permanent and the member states of the current multilateral bodies and state alliances, potential platforms for cooperation and in particular the necessary cooperation against MUAI for limiting conflicts, might be sucked into great power rivalry
Since EU considers itself as complementary to NATO, the danger is that EU member states will be sucked into the great power geopolitical confrontation according to NATO priorities (Exclusive Euro-Atlantist ideology) the « West against Russia and China ». This is reducing the aim of EU for strategic autonomy. Instead of promoting cooperation at global level with all states and nations to face the threats of MUAI and Cognitive warfare, EU member state risk being aligned on NATO specific interests and objectives.
Conclusion : recommendations
Ethics is necessary regarding the rapid development of AI and requires large international cooperation. If EU is aware of ethics, it is demonstrating large contradictions in its positioning to fight against strategic MUAI.
Without a preliminary diagnosis of the problem, it will not be possible to deal with the geopolitical implications of strategic MUAI and its implications for the EU and international psychological security. Although it claims to have achieved more strategic autonomy, the EU is reinforcing de facto the emergence of antagonistic alliances, including those engaged in AI development, by positioning itself as part of an exclusive Euro-Atlantic alliance that resists multipolarity (rather than expounding a doctrine of non-alignment).
A better balance is needed to avoid escalation and uncontrolled spiral of geopolitical rivalries without limits in time and space. On a global scale, there will be no international order and no common rules and norms for the development of AI to fight MUAI without an acceptable geopolitical order involving the Great Powers. There must be an acceptance of an ethical and human-centered order, a new multilateral configuration that offers a model for global AI cooperation. Only then will MUAI and threats to IPS be contained.
Otherwise, strategic MUAI in the context of Great Power rivalry and its threat to IPS will open up a Pandora’s box of world conquest by a new entity. This may lead to total and permanent war and place humanity in danger.
To avoid the polarization of rival blocs of different states and multilateral organizations, is it possible to promote a new multilateral cooperation on AI and MUAI and based on a ethical and human-centered order combined with a better geopolitical balance of power? By promoting new models for global AI cooperation between BRICS, SCO, EU, and other state alliances under coordination and scrutiny of UN, only then will MUAI, including Cognitive war, and threats to IPS can be contained. (Diagram 3)
International cooperation based on inclusiveness, respect and reciprocity will be better achieved with a stronger geopolitical balance on AI between global actors such as the US, China, Russia and the EU member states, and also between smaller states. International actors should therefore place more emphasis on questions regarding geopolitical balance and data sovereignty to counter threats to IPS from MUAI. It should also focus more on the different consequences that it could face regarding strategic MUAI, such as the implications for the EU of Cognitive Warfare and the development of GEOINT that goes beyond tactical MUAI.
Strong bilateral or smaller coalitions should be created for cooperation between voluntary actors who would agree to pool the necessary resources and skills in order to create awareness of dangers of Cognitive Warfare, and ensure their independence and their future digital sovereignty, and to avoid being sucked into the US-China confrontation.
The EU should refrain from aligning itself with potential new and emerging exclusive alliances as a result of the increasing confrontation between the US, China and Russia, but should instead promote strategic autonomy and sovereignty, and cooperation on an inclusive basis. (Diagram 4)
Questions of power and sovereignty should also be geared towards a more socially cohesive and innovation-oriented model of development and a better combination of multipolarity and multilateralism (a better accepted multipolarity is the condition of reinforced multilateralism) for the whole world.
International cooperation is also necessary (under UN/UNESCO coordination) on forward-looking threat assessments, including the mapping of the AI threat landscape associated with Cognitive Warfare and global cooperation in building projects using AI to counter MUAI. An international research centre, on the risk of MUAI and its threat to IPS could be promoted with voluntary states and should not be based on a restrictive membership reflecting the emergence of geopolitical blocs (such as an “alliance of democracies”).
[i] https://digital-strategy.ec.europa.eu/en/policies/regulatory-framework-ai
[ii] Independent administrative authorities may be subject to greater political influence than judicial authorities (Hacker, 2024).
[iii] https://www.linkedin.com/pulse/eu-ai-act-implications-use-countries-national-defense-frederiksen-uvrqe/
[iv] Francois du Cluzel, Cognitive Warfare, Innovation Hub, sponsored by ACT (Allied Command Transformation) of NATO, November 2020, 45 p.
[v] European Parliament (2019) Polarisation and the use of technology in political campaigns and communication. https://www.europarl.europa.eu/thinktank/en/document.html?reference=EPRS_STU(2019)634414.
[vi] Chessen M (2017) The Madcom future: How artificial intelligence will enhance computational propaganda, reprogram human culture, and threaten democracy… and what can be done about it. Atlantic Council. https://www.atlanticcouncil.org/in-depth-research-reports/report/the-madcom-future/.
[vii] European Parliament (2020) Framework of ethical aspects of artificial intelligence, robotics and related technologies, European Parliament Resolution of 20 October 2020 with recommendations to the Commission on a framework of ethical aspects of artificial intelligence, robotics and related technologies (2020/2012(INL). https://www.europarl.europa.eu/doceo/document/TA-9-2020-0275_EN.html.
