Dear colleagues,
It was a pleasure spending three days with you during our course on AI and diplomacy. We hope we achieved our three main goals:
- Gaining a basic understanding of AI (patterns and probability).
- Developing skills to engage in planning and strategic decision-making on AI in your organisations and ministries.
- Enhancing understanding of AI governance and diplomacy.
In this course log, you will find a summary of our learning journey, along with links to additional materials to deepen your knowledge of AI and diplomacy.
We wish you much success in exploring the applications of AI in both your personal and professional life.
With best regards,
Sorina, Andrej, and Jovan
CourseLog
First day: 18 November
We started the course with understanding AI through patterns and probability: We delved into the foundational concepts of AI, using patterns and probability to demystify the technology. We explained AI through the metaphor of national flags as you can see from this video…
We divided deeper into the following aspects of AI:
- Philosophical and critical thinking tools: We outlined philosophical frameworks and critical thinking strategies that enhanced your understanding of AI and its implications.
- AI diplomacy trinity: We examined the intersection of AI and geopolitics, exploring the changing geopolitical landscape, emerging topics on the diplomatic agenda, and innovative tools for effective diplomacy. We dived into the fascinating world of AI geopolitics together!
- AI geopolitics: We focused on the impact of AI on geopolitics. We started with geopolitics around submarine cables which carry 95% of global digital traffic. Without submarine cables there is no digital interaction and AI. Here you can see how the UAE is connected to submarine cables. Below is a video on cable geopolitics…
Follow-up answers
Here are answers to two questions asked:
- A survey of the use of AI in conflict resolution and mediation.
- MIT Moral Machine hows cultural differences in ethics that can impact AI (e.g. whom to save: elderly persons, mothers with children, or two reach busnissmans).
Second day: 19 November
During our second day, we focused on AI geopolitics and AI topics and policy issues. We centered discussion around the recent article from Foreign Affairs Journal: ‘The Emerging Age on AI Diplomacy’.
The Emerging Age of AI Diplomacy
Originally published in Foreign Affairs
Oct 28, 2024
In a vast conference room, below chandeliers and flashing lights, dozens of dancers waved fluorescent bars in an intricately choreographed routine. Green Matrix code rained down in the background on a screen that displayed skyscrapers soaring from a desert landscape. The world was witnessing the emergence of “a sublime and transcendent entity,” a narrator declared: artificial intelligence. As if to highlight AI’s transformative potential, a digital avatar—Artificial Superintelligence One—approached a young boy and together they began to sing John Lennon’s “Imagine.” The audience applauded enthusiastically. With that, the final day dawned on what one government minister in attendance described as the “world’s largest AI thought leadership event.”
This surreal display took place not in Palo Alto or Menlo Park but in Riyadh, Saudi Arabia, at the third edition of the city’s Global AI Summit, in September of this year. In a cavernous exhibition center next to the Ritz Carlton, where Crown Prince Mohammed bin Salman imprisoned hundreds of wealthy Saudis on charges of corruption in 2017, robots poured tea and mixed drinks. Officials in ankle-length white robes hailed Saudi Arabia’s progress on AI. American and Chinese technology companies pitched their products and announced memorandums of understanding with the government. Attendants distributed stickers that declared, “Data is the new oil.”
For Saudi Arabia and its neighbor, the United Arab Emirates (UAE), AI plays an increasingly central role in their attempts to transform their oil wealth into new economic models before the world transitions away from fossil fuels. For American AI companies, hungry for capital and energy, the two Gulf states and their sovereign wealth funds are tantalizing partners. And some policymakers in Washington see a once-in-a-generation opportunity to promise access to American computing power in a bid to lure the Gulf states away from China and deepen an anti-Iranian coalition in the Middle East.
They should temper their expectations. Saudi Arabia’s and the UAE’s economic and political relationships with China are more robust than ever, and that is unlikely to change. Although the Gulf states are eager for advanced AI chips that for now only the United States can provide, they also have strong and enduring incentives to hedge their bets, playing the major powers off against each other to extract concessions. When appropriate, the United States and its tech companies should cooperate with the Gulf states on AI. But they should do so within limits and with safeguards—and without deluding themselves that doing so will bring a lasting strategic realignment in the Gulf.
BRIDGING THE GULF
The two Gulf states’ interest in AI is not new, but it has intensified in recent months. Saudi Arabia plans to create a $40 billion fund to invest in AI and has set up Silicon Valley–inspired startup accelerators to entice coders to Riyadh. In 2019, the UAE launched the world’s first university dedicated to AI, and since 2021, the number of AI workers in the country has quadrupled, according to government figures. The UAE has also released a series of open-source large language models that it claims rival those of Google and Meta, and earlier this year it launched an investment firm focused on AI and semiconductors that could surpass $100 billion in assets under management.
U.S. technology companies have eagerly reciprocated this interest. The infrastructure required to train the latest generation of AI models uses vast amounts of energy, capital, and land—three things the Gulf states have in abundance. OpenAI chief executive Sam Altman has talked with investors in the UAE about multitrillion-dollar investments in chips and data centers, and state-backed Emirati firms participated in OpenAI’s recent round of fundraising. Top executives at the semiconductor giants Taiwan Semiconductor Manufacturing Company and Samsung have floated the idea of building factories in the UAE. Amazon announced a $5.3 billion investment for data centers in Saudi Arabia earlier this year, and the AI startup Groq has partnered with Saudi Arabia’s state-owned oil giant Aramco to build a huge AI data center in the country. Microsoft, meanwhile, has invested $1.5 billion in the UAE’s leading tech company, G42, in a deal that will help Microsoft expand its business in emerging economies and give G42 access to Microsoft computing power.
Where American AI companies see a commercial opportunity, some policymakers in Washington see a strategic one: access to U.S. computing power could be an important carrot to draw countries away from a rapidly expanding Chinese technological ecosystem. The United States wants to shore up its relationship with the world’s largest oil exporters and deepen an anti-Iranian coalition in the Middle East.
Both Saudi Arabia and the UAE are increasingly influential in the region and beyond—in 2023, for example, the UAE announced $45 billion in investments in Africa, far surpassing Chinese expenditures there that year. It is in Washington’s interests that Gulf actors invest their vast sums of capital in U.S. technology companies rather than Chinese ones.
Washington has a good deal of leverage over these technological partnerships because exporting the advanced chips used in AI data centers requires licenses from the U.S. government, which has been slow-walking approvals for large-scale sales for months while it debates what conditions to attach.
If the U.S. government doesn’t greenlight these licenses, some fear, China might soon offer an alternative. At the AI summit in Riyadh, the subject of U.S. export controls was a regular conversation starter.
Google and Microsoft had the most prominent booths by the entrance, but the Chinese firms Alibaba and Huawei were not far away, their booths stationed in an adjoining room around the corner—a tangible reminder of the Chinese options that may be available to the Gulf states if Washington adopts a more restrictive approach.
HEDGING THEIR BETS
Even though the United States has an economic and geopolitical opportunity in the Gulf, there are also significant risks to offshoring major clusters of advanced AI chips to authoritarian regimes with elaborate surveillance systems, an appetite for military adventurism, and expanding ties to China. Lawmakers and Pentagon officials have expressed concern that Chinese companies linked to the People’s Liberation Army could access those chips through data centers in the Middle East as a means of skirting U.S. export controls that have sought to restrict China’s access to cutting-edge AI technology.
More broadly, if AI systems soon gain the potential to drive explosions in economic growth, design new synthetic bioweapons, or develop impressive new cyber-capabilities, they may disrupt the global balance of power. If that proves to be the case, then the infrastructure that underpins frontier AI systems—in particular, the massive data centers where these models will be trained and hosted—should not be offshored lightly. As the former OpenAI researcher Leopold Aschenbrenner put it in a widely circulated memo: “Do we really want the infrastructure for the [next] Manhattan Project to be controlled by some capricious Middle Eastern dictatorship?”
The UAE in particular appears to have made serious efforts to assuage these concerns, going out of its way to portray itself as a responsible steward of American AI technology. According to public reporting, the UAE has pledged that it will lock down its data centers, stripping them of Chinese hardware that might have backdoors, screening customers and workers, and monitoring how buyers use their chips. Under U.S. pressure, G42, which is chaired by the Emirati national security adviser Sheikh Tahnoon bin Zayed, divested from Chinese firms and stripped out its Huawei technology as part of its deal with Microsoft. Last month, partly in response to these efforts, the U.S. Department of Commerce published a rule that could ease the shipment of AI chips to the Middle East.
The UAE has declared that it seeks a “marriage” with the United States founded on AI. But U.S. policymakers should understand that any such marriage is unlikely to be monogamous. Saudi Arabia and the UAE both have powerful incentives to hedge their bets, given American domestic political instability and the enduring, if eternally frustrated, U.S. desire to “pivot” to Asia. China is Saudi Arabia’s largest oil customer and trading partner and the UAE’s top non-oil trading partner. It does not hector either state about its human rights abuses or regional activities. Chinese-made drones are among the UAE’s tools of choice for its covert campaigns in Sudan, and earlier this year the Chinese and Emirati air forces held joint exercises in Xinjiang, of all places. And even though G42 may have divested from Chinese firms, a new Abu Dhabi investment vehicle has taken over the management of G42’s Chinese-focused fund, and, like G42, the new vehicle is overseen by the Emirati national security adviser. At another conference in Abu Dhabi last month, Chinese and Emirati officials alike described the last few years as the “golden era” of Chinese-Emirati cooperation.
MAKE YOURSELF COMFORTABLE
Even in the face of such hedging, the United States should not impose a blanket ban on all sales of advanced AI chips to the Gulf. Many, if not most, emerging powers believe that they can successfully balance relationships with both the United States and China, and U.S. policymakers should generally restrain themselves from pressuring regional powers into making zero-sum choices. At times, U.S. policymakers will have to become comfortable operating in regions and sectors in which U.S. and Chinese influence overlap. And it would not serve U.S. interests if Washington were to drive billions of dollars of Gulf funds toward projects that accelerate China’s technological progress.
U.S. policymakers should thus move forward with their negotiations with the Gulf states over chip exports. But they should do so without any illusions about the regimes they are working with, the risks involved, or the chances that such collaboration will help reshape the political order of the Middle East. The Gulf states will not cut off ties with China except in narrowly scoped areas, and even then such decisions will always be open to renegotiation.
Without serious efforts at mitigation in the form of sustained investments in both physical and cybersecurity, building massive data centers in non-allied countries increases the risks of intellectual property theft and misuse, especially if those centers host the weights of frontier models (the parameters that encode the core intelligence of an AI system).
The United States will need to devote resources to monitor—and enforce—compliance for any deals it reaches. In the absence of independent verification, the United States should treat Emirati and Saudi assurances about their stewardship of U.S. technology with skepticism. And U.S. policymakers should strongly encourage American tech companies to build their largest and most advanced facilities in the United States.
In this emerging era of AI diplomacy, Washington will face similar challenges in one setting after another: it will have to control the proliferation of technologies that might have critical national security implications without kneecapping American corporations or driving potential partners into the arms of China. In their negotiations with the Gulf, U.S. policymakers should make sure that they set the right precedents.
In the next step, we discussed AI governance by looking at a simple question: Who can answer calls for AI and digital solutions by citizens, companies, and countries worldwide?
All AI governance issues can be located on the 4 layers of AI Governance Pyramide…
Here are 158 Sorina’s slides on AI governance and diplomacy…
Third day: 20 November
During the third day, Sorina continued with a summarisation of AI governance. After that, we explained the negotiation exercise on the Lethal Autonomous Weapons Systems (LAWS), or more colloquially, ‘Killer Robots’.
For more details on past and ongoing negotiations on LAWS at the UN level, you can consult our page on the dedicated Group of Governmental Experts.
Negotiation exercise: Lethal Autonomous Weapons System
The simulation exercise includes five AI assistants for negotiations:
The simulation exercise focuses on negotiations on a more precise formulation for human control of ‘killer robots’.
Text to be negotiated:
Guiding Principles affirmed by the Group of Governmental Experts on Emerging Technologies in the Area of Lethal Autonomous Weapons System
It was affirmed that international law, in particular the United Nations Charter and International Humanitarian Law (IHL) as well as relevant ethical perspectives, should guide the continued work of the Group. Noting the potential challenges posed by emerging technologies in the area of lethal autonomous weapons systems to IHL, the following were affirmed, without prejudice to the result of future discussions:
(a) International humanitarian law continues to apply fully to all weapons systems, including the potential development and use of lethal autonomous weapons systems;
(b) Human responsibility for decisions on the use of weapons systems must be retained since accountability cannot be transferred to machines. This should be considered across the entire life cycle of the weapons system;
(c) Human-machine interaction, which may take various forms and be implemented at various stages of the life cycle of a weapon, should ensure that the potential use of weapons systems based on emerging technologies in the area of lethal autonomous weapons systems is in compliance with applicable international law, in particular IHL. In determining the quality and extent of human-machine interaction, a range of factors should be considered including the operational context, and the characteristics and capabilities of the weapons system as a whole;
(d) Accountability for developing, deploying and using any emerging weapons system in the framework of the CCW must be ensured in accordance with applicable international law, including through the operation of such systems within a responsible chain of human command and control;
(e) In accordance with States’ obligations under international law, in the study, development, acquisition, or adoption of a new weapon, means or method of warfare, determination must be made whether its employment would, in some or all circumstances, be prohibited by international law;
(f) When developing or acquiring new weapons systems based on emerging technologies in the area of lethal autonomous weapons systems, physical security, appropriate non-physical safeguards (including cyber-security against hacking or data
spoofing), the risk of acquisition by terrorist groups and the risk of proliferation should be considered;
(g) Risk assessments and mitigation measures should be part of the design, development, testing and deployment cycle of emerging technologies in any weapons systems;
(h) Consideration should be given to the use of emerging technologies in the area of lethal autonomous weapons systems in upholding compliance with IHL and other applicable international legal obligations;
(i) In crafting potential policy measures, emerging technologies in the area of lethal autonomous weapons systems should not be anthropomorphized;
(j) Discussions and any potential policy measures taken within the context of the CCW should not hamper progress in or access to peaceful uses of intelligent autonomous technologies;
(k) The CCW offers an appropriate framework for dealing with the issue of emerging technologies in the area of lethal autonomous weapons systems within the context of the objectives and purposes of the Convention, which seeks to strike a balance between military necessity and humanitarian considerations.
AI Assistants
We developed three AI assistants to support course discussion and coverage…
UAE AI and Digital Governance
Learn more about this AI Assistant
It is developed by DiploAI by using the following inputs:
Annotated texts of AI and digital governance and policy worldwide
Annotated academic and research papers
UAE’s policy and regulatory materials, including:
- Internet access management policy (2017)
- International telecommunications cable regulations (2022)
- UAE’s 4th industrial revolution strategy (2017)
- Federal Decree-Law No. (34) of 2021 On Countering Rumors and Cybercrimes
- National Cloud Security Policy (2023)
- National Policy for the Internet of Things Security (2023)
- Science, Technology and Innovation Policy (2024)
- General Framework for Adopting Sustainable Digital Transformation (2024)
- Digital Government Service Policy (2023)
- Critical information infrastructure protection policy (2023)
- National Information Assurance Framework (2023)
- UAE Information Assurance Regulation (2020)
- UAE Digital Government Strategy 2025 (2023)
- Federal Decree by Law no. 45 of 2021 concerning the protection of personal data (2021)
AI House of Wisdom
Learn more about this AI Assistant
It is developed by DiploAI by using the following inputs:
Annotated texts of AI and digital governance and policy worldwide
Annotated academic and research papers
UAE’s policy and regulatory materials, including:
During the Islamic Golden Age, spanning from the 8th to the 14th century, scholars produced numerous seminal works across various disciplines. Here is a selection of notable original books and materials from that era:
“Kitāb al-Ḥiyal” (The Book of Ingenious Devices) by Banū Mūsā brothers (9th century CE): A compilation of mechanical devices and automata, showcasing early engineering ingenuity.
“Kitāb al-Fihrist” (The Book Catalogue) by Ibn al-Nadim (987 CE): A comprehensive bibliographic compilation detailing the knowledge and literature of 10th-century Islam, referencing approximately 10,000 books and 2,000 authors. Wikipedia
“Al-Muqaddimah” (The Introduction) by Ibn Khaldun (1377 CE): A pioneering work in historiography and sociology, offering profound insights into the philosophy of history and the rise and fall of civilizations.
“Kitāb al-Manāẓir” (The Book of Optics) by Ibn al-Haytham (1021 CE): A foundational text in optics and physics, introducing the scientific method and significantly influencing the development of modern science.
“Al-Qanun fi al-Tibb” (The Canon of Medicine) by Ibn Sina (Avicenna) (1025 CE): An encyclopedic medical text that served as a standard reference in both the Islamic world and Europe for centuries.
“Kitāb al-Jabr wa-l-Muqābala” (The Compendious Book on Calculation by Completion and Balancing) by Al-Khwarizmi (9th century CE): The seminal work that introduced the fundamental principles of algebra.
“Kitāb al-Hayawān” (The Book of Animals) by Al-Jahiz (9th century CE): An extensive treatise on zoology, discussing animal behavior, classification, and the concept of natural selection.
“Kitāb al-Buldan” (The Book of Countries) by Al-Ya’qubi (9th century CE): One of the earliest works in Islamic geography, providing detailed descriptions of various regions and cultures.
“Al-Kāmil fī al-Tārīkh” (The Complete History) by Ibn al-Athir (1231 CE): A comprehensive history of the world from creation up to the 13th century, focusing on Islamic history.
“Risāla fī Istikhrāj al-Mu’āmalāt” (Treatise on the Extraction of Measures) by Al-Karaji (10th century CE): A significant work in mathematics, particularly in algebra and arithmetic.
Energy, environment, and climate in AI models and data processing
Relevant publications
Here are publications of direct relevance for our course discussion on AI and diplomacy.
Unpacking Global Digital Compact
Unpacking the Global Digital Compact explores the origins and impact of the Global Digital Compact (GDC), a new UN framework adopted at the 2024 Summit of the Future to shape digital and AI governance. This book explores the GDC’s complex negotiations and its role in fostering global cooperation amid diverse perspectives and diplomatic challenges. READ MORE
Diplomacy Reimagened: Competencies 2040
The book explores the evolving importance of talents, knowledge, and skills, emphasising the need for new competencies and continuous learning in the AI-driven era. Drawing from decades of experience in diplomatic training, the book offers insights into how the profession must adapt to thrive in this rapidly changing landscape. READ MORE
Understanding AI througn national flags
Understanding how AI functions is becoming necessary for everyone, from citizens to societal leaders. AI poses two levels of explainability challenges: one is for technical people to understand how neural networks function, and the other is for the wider population to understand AI and what it is capable of. READ MORE
History of Diplomacy and Technology
The book ‘History of Diplomacy and Technology’ reminds us that every new technology throughout history has promised to change the way diplomacy is conducted. Some changes occurred, but the essence of diplomacy remained the same: the peaceful resolution of disputes through negotiation and mediation. This historical journey of diplomacy’s continuity and technological changes since smoke signals is thought-provoking reading as we consider the future of diplomacy in the AI era. READ MORE
Tech Diplomacy: Actors, Trends, and Controversies
In today’s world, tech diplomacy bridges governments and tech companies, focusing on governance, policy, and cooperation in digital technologies and AI. This publication examines its definition, relevance, key actors, methods, and global hubs. It builds on prior reports and highlights Denmark’s pioneering efforts in establishing a dedicated tech diplomacy policy. READ MORE
An Introduction to Internet Governance
‘An Introduction to Internet Governance provides an excellent entry point. It has introduced many diplomats and officials to this emerging field of global policy. For others, it will stimulate reflections from linguistic, legal, and other perspectives. This book clearly shows that although the Internet is a ‘technical’ invention, its governance is far from simply a technical issue. Kurbalija’s book highlights the legal, social, linguistic, and economic perspectives of Internet governance. It is an impressive introduction to this emerging field of global policy’ Nitin Desai, former Chair of the Internet Governance Forum (IGF). READ MORE
Questions and Answers
During our visit to the UAE, we had many stimulating discussions with students, researchers, and diplomats. Here is a survey of questions and answers from the visit: