Computational diplomacy, for one, is concerned with our emerging ability to map the landscape of international relations, to gather and analyse data on unprecedented scales and to simulate potential outcomes. This has transformational potential for diplomatic activity. For instance, efforts have already begun to plot the networks of influence between actors on an international scale and to use artificial intelligence to mine the large databases of texts relating to historical negotiations and international organisations. As such, computational diplomacy is revealing not only the complexity of modern international relations but the potential knock-on effects of future actions, giving opportunities to favour the emergence of desired outcomes. It also allows actors to better understand the history of negotiations and how changes in language reveal movements in position, and to reduce uncertainty in formulating plans.
Negotiation engineering, on the other hand, is a solution-oriented approach to negotiation problems that uses quantitative methods in a heuristic way to find an adequate solution. The approach draws on breaking down and formalising the problem(s) at hand and on the heuristic application of mathematical methods such as game theory and mathematical optimisation. In this way, it can contribute to recognising feelings and better managing them during the negotiation process, as well as allowing for resolutions of more complex real-world issues.
KEY TAKEAWAYS
Researchers from a wide range of academic disciplines are collaborating to improve approaches to diplomacy through applying numerical and computational techniques as well as AI. Computational diplomacy, for instance, can exploit existing data to better understand past diplomatic achievements and the network structures that facilitated them. Also of interest is Negotiation engineering, which is being transformed by AI techniques and raises the possibility that future agreements will be informed by computationally derived understanding and will more successfully bring together broader groups of stakeholders in complex negotiations, while allowing progress with fewer missteps. Similarly, Predictive peacebuilding has potential to use machine-learning approaches to increase the chances of achieving greater international stability. Trust and cooperation modelling applies computational approaches to the task of establishing relationships that result in positive experiences of cooperation and collaboration experiences while providing tools for distinguishing trustworthy from untrustworthy potential partners.
Three fundamental questions guide GESDA’s mission and drive its work: Who are we, as humans? How can we all live together? How can we ensure the well-being of humankind and the sustainable future of our planet? We asked researchers from the field to anticipate what impact future breakthroughs could have on each of these dimensions. This wheel summarises their opinions when considering each of these questions, with a higher score indicating high anticipated impact, and vice versa.
The cost of processing this data means it has not been well used to inform the process of diplomacy, to amplify cooperation and to improve outcomes. Nevertheless, organisations like the UN, the World Bank and other policy-makers are working hard to integrate quantitative methods into their organisations, which will accelerate the practice of computational diplomacy and its use of big data, machine learning and computational thinking.
There is much low-hanging fruit here. The networks of actors on the international stage and their institutional relationships are already beginning to be mapped,1,2,3 giving a deeper understanding of the connections that can influence negotiations. Also being mined for insight are resolutions adopted by various international organisations. Relevant examples include the United Nations General Assembly and Security Council resolutions histories,4 sponsorship data5 and debate themes, as well as the resolutions adopted by the World Health Organization.6,7 This work is beginning to deliver quantitative results, a crucial step for reproducibility.8,9,10
Much more can be done. Combining a data-driven approach with computational modelling can facilitate a more fundamental understanding of how multilateral governance systems work and how they can be improved, for example.11 The growing use of AI is likely to have a significant impact here,12 but developing the multidisciplinary expertise that can manage and exploit diplomatic processes is a significant challenge.
This approach could help to reveal the ingredients of success for global cooperation. Numerous non-governmental organisations in complex networks currently aim to solve complex problems such as climate change and the challenges related to Sustainable Development Goals. And yet they often achieve far less than they hope. A better understanding of the way organisations should interact to achieve specific goals could change this.
Negotiation engineering has already achieved a number of practical successes. For instance, in the diplomatic sphere, the approach played a crucial role in the land transport agreement between Switzerland and the European Union, and in facilitating nuclear talks between Iran and the P5+1 group of nations.14
Negotiation engineering does not intend to replace face-to-face discussion and neither does it seem likely to ever do so. It may in some cases also have limited application: not all problems are quantifiable or should be reduced to a quantitative level. However, in case a negotiation involves problems with a particular degree of complexity and actors with a certain level of analytical capacity open to a rational approach, negotiation engineering can allow for more logical accuracy in finding pragmatic solutions while building trust between mediators.
The availability of negotiation tools powered by AI has triggered a transformative period in negotiation engineering. These tools are helping human negotiators become more effective and has kick-started a new era of automated negotiations with their own strategies and characteristics. This evolution is paving the way for a greater exploration of the strategy space and the development of new negotiation theories that encompass and extend established theory.15,16
The field has been bolstered by a number of successes, particularly with the application of AI techniques.18 For example, machine-learning-based analysis of newspaper text can predict the onset of conflict, becoming particularly useful when risk in previously peaceful countries arises.19,20 Analysis of food prices shows that increases beyond a threshold level are correlated with civil unrest in many parts of the world.21 Granular, actor-based conflict data22,23 gives rise to models that take it into account.24,25 The ability to model migration patterns on a global scale using anonymised Facebook data is also a significant step.26
Research in the field sees policies for conflict prevention — mediation, development aid, institutional reform and building state capacity, for example — as a prediction policy problem.27 This means that the treatment effects of different policies and the targeting of these policies in time and space both need to be studied quantitatively. However, important limitations and potential pitfalls remain. Current conflict models have limited ability to make causal inferences and are sometimes informed by outdated data-gathering.
Even more serious is the possibility that predictive models can be self-fulfilling or self-defeating. For example, a prediction of war could cause local populations to flee, raising tensions that themselves trigger conflict. Also, care is needed in defining peace and ensuring that “peaceful” outcomes make a difference to the lives of real people such as refugees and those who have experienced trauma. Understanding these kinds of questions requires granular data gathered on a vast scale to inform decision-making and the cost-effective use of resources.
This has been applied to a wide range of problems, ranging from information-routing algorithms to online search rankings to recommendation algorithms. But there is a broader sense in which trust and cooperation studies are useful — in modelling the way people behave in the groups that make up societies.28
In any society, business or network, the ability to evaluate and then trust partners is a crucial component of cooperation.29 Applying trust modelling to the networks of actors at work in the diplomatic landscape has the potential to better model potential outcomes of discussions, votes and negotiations.
This work comes at a time when the role of trust in broader society has been thrown into sharp focus by the phenomenon of fake news, manipulated images and deepfake videos. The diplomatic landscape is powerfully shaped by the information that flows through it, and false and misleading information has huge disruptive potential.30
A key emerging issue is the role of AI and how it will be used to understand and inform multilateral decision-making processes. At the heart of this question is whether AI systems will become better at interpreting the complex data fed into them or worse as AI-generated data distorts their view of the world. The possibility that AI systems could create a kind of artificial truth will be an important issue for the field.
