AI systems are evolving rapidly, and with them, the urgency of their governance across domains and disciplines. From deepfakes and AI-enabled cyber operations to increasingly autonomous systems in the military domain, AI has long ceased to be a frontier question and is already re-shaping national, regional and global security landscapes.
As advances in AI raise increasingly complex questions for policymakers, regulators and international institutions alike, States and stakeholders are seeking shared understandings, common norms and coordinated approaches that can support responsible governance across security and defence contexts.
This growing momentum is reflected in an expanding ecosystem of international and regional initiatives. Recent developments, including the UN General Assembly resolutions 79/239 and 80/58, the Secretary-General’s report on AI in the military domain, REAIM and a growing number of national and regional AI strategies, point to a broader shift from abstract debates toward implementation, practical governance measures and sustained multilateral engagement.
At the same time, the pace of technological innovation continues to challenge existing policy frameworks, making international cooperation and cross-sector dialogue increasingly necessary.
Amid this rapidly evolving landscape, UNIDIR’s Global Conference on AI, Security and Ethics 2026 (AISE26) serves as an agile and inclusive space for bridging technical and policy communities and strengthening cooperative approaches to AI-related security challenges.
A platform for dialogue
The inaugural edition of the conference took place in March 2025, against a highly dynamic backdrop. Policymakers and regulators worldwide had begun to recognize the urgency of shared understandings, norms and regulations that transcend national borders and individual interests, including in the context of AI’s applications for security and defence.
In parallel, AI in the military and security domains were often, if not most of the times discussed at either end of the spectrum: as a question of governance, international law and global norms, or as a question of technical design and corporate governance.
In response, UNIDIR leveraged its existing network of experts – building on the work of the Roundtable for AI, Security and Ethics (RAISE) launched in 2024 – and giving a voice to under-represented perspectives from all five UN regional groups in an inclusive manner that transcends geopolitical barriers.
As such, UNIDIR’s Global Conference on AI, Security and Ethics constituted a platform for independent, neutral and expert-led dialogue.
By fostering mutual understanding and cross-sectoral action, it established itself as a channel for substantive input feeding into multilateral, international, regional and national governance initiatives. The conference was also held ahead of the deadline for the submission of States’ and the multistakeholder community’s views to the UN Secretary-General pursuant to resolution 79/239 on AI in the military domain and its implications for international peace and security.
Unwavering commitment to AI and security
UNIDIR’s conference is also part of the Institute’s broader work on AI and autonomy. This explores these technologies’ evolution, their applications and implications on international, regional and national peace and security, and the policy options available to respond effectively – informing both the deliberations on lethal autonomous weapons systems and broader discussions on military AI.
Recent projects include the co-development with OHCHR of a voluntary framework that provides guidelines for industry actors in AI in the military domain. Acknowledging the private sector’s increasingly important role throughout these technologies’ lifecycle, UNIDIR seeks to collaborate with organizations, industry actors, civil society and Member States to formulate guidelines translating existing international law and norms into measures for their implementation and operationalization.
The Institute emphasizes regional engagement for the responsible development, deployment and use of AI in the military domain and for the application of international humanitarian law in relation to lethal autonomous weapons systems. It does so while preserving an inherently inter-disciplinary approach due to the multi-faceted nature of these technologies. This is reflected in its work surrounding large language models, decommissioning, as well as the publication of its roadmap encapsulating the Institute’s recommendations for future policy action.
Additionally, given the growing number of national initiatives in this space, UNIDIR has developed, over the past years, tools for furthering national capacity. These include its AI Policy Portal, which fosters transparency and information-sharing, and is complemented by the Draft Guidelines for the Development of a National Strategy on AI in Security and Defence. In addition, UNIDIR’s Women in AI Fellowship, taking place annually in Geneva, equips women diplomats with the expertise to explore the latest and emerging AI applications from policy, legal and technical lens.
What’s next for AI, security and ethics
UNIDIR’s Global Conference on AI, Security and Ethics 2026 will convene policymakers, diplomats, military representatives, industry leaders, researchers, academia and civil society to address the evolving implications of AI for international security and resilience.
Taking place on 18–19 June at the Palais des Nations in Geneva, and online, the conference serves as a catalyst for action at a pivotal moment for global AI governance.
By bringing together technical and policy communities from across sectors and regions, the conference reflects a shared commitment to a more secure world. Through an expanded programme featuring technical deep dives, high-level panels, interactive roundtables, demonstrations and poster presentations, it seeks to create space not only for dialogue across disciplines, but also for practical exchange between technical and policy communities.
In the lead-up this year’s edition, UNIDIR is also expanding these discussions through the Road to AISE26 webinar series and the launch of the AISE Markers policy notes. Together, these initiatives aim to sustain dialogue ahead of the event, while helping to identify emerging questions, areas of convergence and issues that continue to require deeper reflection across policy and technical communities.
At a time when technological developments risk outpacing governance processes, spaces for inclusive, informed and action-oriented dialogue are critical. The governance of AI and its implications for international peace and security cannot be addressed in isolation. It requires sustained cooperation, shared understanding and continued engagement across the international community in the years ahead.
