Legislators at Lisbon Conference stress the importance of forward-looking regulations to mitigate AI-related security challenges


LISBON, 20 May 2024 – Governments, parliaments, and international organizations must enhance co-operation and develop robust, comprehensive policies to address the multifaceted challenges posed by Artificial Intelligence (AI), said participants today at the 2nd Lisbon Conference of the OSCE PA. Organized by the Portuguese Parliament (Assembleia da República) and co-hosted by the OSCE Parliamentary Assembly, the event aimed to raise awareness about AI's profound security implications, facilitate dialogue among policymakers and key stakeholders, as well as to explore effective policy frameworks and international co-operation avenues to mitigate AI-related threats.

emb img Lisbon1Family Photo, Lisbon, 20 May 2024The conference, titled “Security in the Age of Artificial Intelligence,” brought together around 170 participants, including 70 parliamentarians from 27 OSCE participating States and numerous experts from academia,international bodies, such as the European Parliament and the Council of Europe, and the private sector. 

"It is essential that Parliamentarians play a pivotal role in shaping AI policies that strike a balance between innovation and security, guided by ethical principles and democratic processes," said President Pia Kauma (Finland) in her opening remarks. “AI is a priority within our Assembly,” she added, citing the recent debate of the OSCE PA General Committee on Economic Affairs, Science, Technology and Environment at the Assembly’s Winter Meeting and the activities and discussions of the Ad Hoc Committee on Countering Terrorism.

José Pedro Aguiar-Branco, President of the Assembleia da República of Portugal, highlighted the need to stay ahead  of AI's rapid advancements. “As profound changes unfold quickly across many domains of our daily lives, it is urgent that parliaments address this issue,” he said. “The decision by the OSCE PA to dedicate the second Lisbon conference to 'Security in the Age of Artificial Intelligence' underscores the significance of these questions in fulfilling the core mission of the OSCE: enhancing co-operative security by promoting democratic values, the rule of law, and sustainable development.”

The opening session also featured a speech by Head of the Portuguese Delegation to the OSCE PA Paula Cardoso and a prerecorded video-message by Helga Maria Schmid, OSCE Secretary General.

The conference included three main sessions. Chaired by Gudrun Kugler (Austria), Rapporteur of the OSCE PA second committee, the first session delved into the threats stemming from AI, emphasizing the challenge of finding common standards  at global, regional, and local levels while balancing innovation with safety. Concerns over digital manipulation, disinformation, privacy and copyright infringements, swarming technology, mass jobs displacement, autonomous decision-making, as well as brain-computer interaction were highlighted, urging the development of mechanisms for independent monitoring  and  coherent policy-making.Participants emphasized the need for a unified global approach to AI governance.

emb img Lisbon22nd Lisbon Conference of the OSCE PA, Assembleia da República of Portugal, 20 May 2024During the second session, chaired by Lucie Potuckova (Czech Republic), Chair of the OSCE PA General Committee on Democracy, Human Rights and Humanitarian Questions, panellists and participants discussed AI's dual role in enhancing democracy and development while highlighting potential pitfalls. The debate focused on the need for coherent global regulations to prevent misuse of AI and uphold human rights, as seen in the recent Council of Europe's Framework Convention on AI - the first international  treaty in this domain. They underscored the importance of transparency, accountability, education, and international co-operation to harness AI's benefits while mitigating risks such as corruption and inequalities. Participants praised the transformative impact of AI on healthcare but also raised concerns about surveillance and ownership, underscoring the need to ensure responsible AI use and safeguard democratic processes

Moderated by Azay Guliyev (Azerbaijan), Chair of the OSCE PA General Committee on Economic Affairs, Science, Technology and Environment, the last session explored the regulatory challenges surrounding AI. Participants and panellists examined the EU's approach, learnt more about the recently adopted Council of Europe's Convention, and shared national experiences. They highlighted the pivotal role of national parliaments in navigating AI's regulatory complexities, stressing the importance of  adopting forward-looking regulations which need to be, and remain, fit-for-purpose in a world which is changing very fast.. They advocated for a cohesive global approach, emphasizing the necessity for synchronized efforts among all key stakeholders.

The conference concluded with remarks from Portugal's Minister of State and Foreign Affairs Paulo Rangel, Deputy Head of the Portuguese Delegation to the OSCE PA Luís Graça, Vice-President of the Assembleia da República and Vice-President of the NATO PA Marcos Perestrello and OSCE PA Secretary General Roberto Montella. 

Montella highlighted the exceptional track record of the OSCE PA in addressing security implications of AI. “Looking ahead, we must recognize the need for continued dialogue and collaboration on AI-related issues within the OSCE framework, possibly also through the creation of new structures which can continue to explore and offer recommendations in this fast-evolving field,” he said.

The Lisbon conference has set the stage for continued discussions at the OSCE PA’s upcoming Annual Session in Bucharest, where recommendations in the field of AI governance and security are expected to be adopted.


Photos of the conference are available for public use on Flickr.



Nat Parry

Head of Communications and Press

Office: +45 33 37 80 55
Mobile: +45 60 10 81 77
Email: [email protected]

  • Facebook
  • twir
  • in
  • inst
  • two
  • video