Agentic AI in an Emerging New World Order: What’s at Stake for Middle Powers and Small States?

On 27 January 2026, the Konrad-Adenauer-Stiftung (KAS) Regional Programme Political Dialogue Asia, in cooperation with TUM Asia and the TUM Think Tank, hosted a seminar titled “Agentic AI in an Emerging New World Order: What’s at Stake for Middle Powers and Small States?” The event explored the socio-political, legal, and technical dimensions of agentic AI, focusing on its potential risks and opportunities for middle powers and small states.

Following the opening remarks by Andreas Klein, Director of the Regional Programme Political Dialogue Asia, the keynote presentation was delivered by Dalia Yousif Ali, PhD candidate, lecturer, and research associate at the TUM School of Social Sciences and Technology. In her presentation, she discussed the implications of agentic AI for countries in the Global South. Dalia explained to the audience the difference between generative AI models, such as ChatGPT, and agentic AI systems that can act and make decisions autonomously. She argued that the key challenge for middle powers and small states is not access to AI, but limited control over the regulatory frameworks governing it. In Dalia’s view, weak monitoring and governance capacity rather than a lack of infrastructure is the main constraint.

The second presentation, “Agentic AI in the Philippines: Opportunities, Risks, and Pathways Forward,” was delivered by Kris R. Villanueva-Libunao, former Deputy Minister of the National Security Council of the Republic of the Philippines and Executive Director of SmartCT. She presented the Filipino perspective and highlighted potential user benefits, risk and opportunities of agentic AI within the ASEAN region. She also addressed the geopolitical risks of being caught between China and the United States, as well as the challenges of establishing and enforcing effective regulatory frameworks. In conclusion, she outlined that there are potential pathways forward, demonstrating that the Philippines and similar states do have viable prospects for achieving AI sovereignty.

The final presentation was delivered by Dr Elena Pribytkova, Senior Research Fellow at the NUS Centre for International Law, under the title “Agentic AI and Digital Colonialism: Can International Law Protect Middle Powers and Small States?” Approaching the issue from the perspective of international law, she examined digital colonialism as a form of domination and discrimination exercised through control over software, hardware, and digital networks. She further stressed that the concentration of power in agentic AI lies not only with states, but particularly with transnational corporations (TNCs) and other non-state actors. She concluded that AI governance under international law requires both the digital independence and digitalisation of international legal frameworks, while human rights standards can provide a “digital minimum” to protect middle powers and small states from digital exploitation.

Following the presentations, Anish Mishra, Project Manager and Research Fellow for Foreign Affairs and Security Policy at KAS Singapore, moderated a discussion with the three speakers. The conversation explored potential regulatory approaches, including those similar to frameworks in the European Union (EU), as well as broader geopolitical dynamics—particularly in relation to the United States and China. The discussion also examined the regulation of both state and non-state actors in the development and deployment of agentic AI systems. The panel also discussed the EU–Indo-Pacific Digital Partnership, and participants agreed that it should be expanded to include more Southeast Asian countries, such as the Philippines, Vietnam, and Malaysia.

The event concluded with a Q&A session and closing remarks by Dr Markus Wächter, Vice President and Executive Director of TUM Asia. The seminar demonstrated that agentic AI is not merely a technological innovation, but also a phenomenon that poses significant geopolitical and legal challenges. For middle powers and small states in particular, agentic AI offers new opportunities while also entailing considerable risks. AI sovereignty emerged as a defining issue of our time and a central theme throughout the seminar.