Search

Begin New Search Topic Emerging Technology Publishing Institution The Geneva Centre for Security Policy

1. Securing AI-based Security Systems

Author:
Sandra Scott-Hayward
Publication Date:
06-2022
Content Type:
Working Paper
Institution:
The Geneva Centre for Security Policy
Abstract:
Fundamental weaknesses of AI include brittleness, embedded bias, catastrophic forgetting and lack of explainability. Although research is under way to address some of these issues, the adoption of AI techniques and models in security systems exposes potentially critical security systems to weaknesses/vulnerabilities such as these. Adversarial training is one strongly recommended approach to increase the robustness (i.e. reduce the brittleness) of the AI model. In this approach, the training dataset is extended to include adversarial examples representative of potential attacks on the system. However, the implementation of adversarial training is currently ad hoc. Given the evidence of AI weaknesses, the omission of adversarial training and similar hardening techniques for AI-based security systems is unacceptable. Standardised testing and evaluation of AI-based security systems is recommended. From a governance perspective, evidence of adversarial robustness evaluation should be a minimum requirement for the acceptance of an AI-based security system. The production of strong adversarial samples does not account for “black swan” events, i.e. random and unexpected events that have an extreme impact. Given that security systems tend to be designed to detect “old” or “known” types of attack, ways need to be found to manage the occurrence of “new” attacks.
Topic:
Security, Science and Technology, Artificial Intelligence, and Emerging Technology
Political Geography:
Global Focus

2. Commitment to Control Weaponised Artificial Intelligence: A Step Forward for the OSCE and European Security

Author:
Anna Nadibaidze
Publication Date:
01-2022
Content Type:
Policy Brief
Institution:
The Geneva Centre for Security Policy
Abstract:
Recent technological and political developments in OSCE participating States suggest a strong interest in pursuing, testing and using weaponised AI and weapons systems with increasingly autonomous features controlled by algorithms. In May 2021 Defence Minister Sergei Shoigu announced that Russia had begun producing combat robots “capable of fighting on their own”,1 while the French Army is planning to introduce robotic systems by 2040.2 The United Kingdom (UK) government has stated its objective of achieving “a leading role in critical and emerging technologies”3 and has established a Defence Artificial Intelligence and Autonomy Unit to better understand them.4 In the United States, the National Security Commission on Artificial Intelligence has urged the government “not [to] be a witness to the AI revolution in military affairs”.5 The global discussion about autonomous weapons systems is often framed in a futuristic way and focuses on lethal autonomous weapons systems (LAWS) – colloquially called “killer robots” – or the “AI arms race”. But weaponised AI is already a reality of European security. Thus far participating States have been reluctant to utilise the OSCE platform to address the risks caused by the increasing autonomy of weapons systems. Building on this issue, this essay intends to address the following questions: (1) how does the lack of regulation of weaponised AI affect security and stability in Europe? and (2) what role can the OSCE play in mitigating the risks related to weaponised AI?
Topic:
Science and Technology, Military Strategy, Artificial Intelligence, and Emerging Technology
Political Geography:
Global Focus

3. Military Operations and Artificial Intelligence

Author:
Tobias Vestner
Publication Date:
07-2021
Content Type:
Working Paper
Institution:
The Geneva Centre for Security Policy
Abstract:
Artificial intelligence (AI) systems will most likely transform military operations. This paper explores how AI systems may affect and be affected by principal instruments for preparing and conducting military operations. Therefore, the paper analyses and discusses AI in the context of strategy, doctrine, plans, rules of engagement, and orders to situate opportunities, challenges, and open questions as well as offer overarching observations. The paper takes a broad angle of analysis that enables a general examination of the issue based on new policies and technological developments as well as the consideration of political, military, legal, and ethical perspectives. Thereby, the paper provides insights and avenues to advance further reflection, research, and policy-making on the appropriate integration, management, and use of AI for military operations.
Topic:
Science and Technology, Military Strategy, Artificial Intelligence, and Emerging Technology
Political Geography:
Global Focus

4. Policy and Technology for the Greater Good: Imagining Futures

Author:
Maricela Munoz
Publication Date:
12-2021
Content Type:
Policy Brief
Institution:
The Geneva Centre for Security Policy
Abstract:
When reflecting on emerging technologies, policy, and participatory perspectives, one may conclude that security must be collective and human-centred. The concept of human security represents a departure from traditional orthodox positions, which focus on the security of the state.40 The subjects of the human security approach are individuals, and the end goal of this approach is the protection of people from all sorts of threats, e.g. disease, poverty, environmental degradation, military action, terrorism, misinformation, etc. Moving the security agenda beyond state security does not mean replacing it, but rather complementing and building on it. Central to this approach is the understanding that a lack of human security can undermine peace and stability within and between states, while an overemphasis on state security can be detrimental to human welfare.41 Science cannot solve all of humankind’s problems, and new groundbreaking discoveries may even endanger human beings. Modern policymaking, based on multidisciplinary and comprehensive participatory methodologies, must be the vehicle to achieve common understandings, reconciling technology for the greater good with political and business priorities. Good implies that it will solve social and environmental challenges. In recent decades public policy has centred on the promotion of macroeconomic growth, but has done little to guarantee individual and societal well-being.42 But it is possible to incorporate public values into policymaking, improving its effectiveness and increasing trust in institutions while informing and educating the general public. Public meetings, multisectoral engagements, external advisory committees, and mediation processes have shown their ability to increase the quality of decision-making and produce a positive impact. The COVID-19 pandemic allowed us to create a new model for the integration of science, technology and policy responses. We must draw important lessons from this crisis, refine the model and create incentives to mainstream good practices. The erosion of multilateralism, nationalist posturing, and the emergence of new actors and powers are changing the global order. International cooperation is failing.43 This could be a pivotal opportunity for a wider societal response, where polymath approaches, strengthened communication, inclusive dialogue and renewed trust shape a new social contract. Today more than ever leaders in communities, countries, and globally must take a proactive approach and call for open and horizontal conversations on the intersection between science and policy. As part of these conversations, all of us are called on to become humanist leaders wherever we are, and to act without delay on the knowledge and understandings that are being built. The meaningful and effective participation of women, youth, indigenous peoples, persons with disabilities, the LGTBIQ+ community, and others is essential to building partnerships at all levels. The current crises we are facing and the expansive modernisation of technologies for a huge range of purposes have underlined the urgency of the need to embrace our human and collective security, protect our fate as a species, and imagine futures that are safe, resilient, and good for all
Topic:
Globalization, International Cooperation, Science and Technology, Humanitarian Intervention, and Emerging Technology
Political Geography:
Global Focus