On 3 October 2025, the Parliamentary Assembly of the Council of Europe adopted Resolution 2628 (2025) on the use of artificial intelligence in migration, asylum, and border management. While acknowledging that AI can improve efficiency, search and rescue operations, and access to information for migrants and refugees, the Assembly stressed that technological innovation must not come at the expense of fundamental rights.

The Assembly cautioned that AI systems, if poorly designed or insufficiently regulated, could reinforce discrimination, undermine privacy, and weaken asylum protections. It underlined that AI should support, but never replace, human decision-making in migration and asylum procedures and called for strong transparency, accountability, and human oversight.

The resolution opposes the use of automated credibility assessments, emotion recognition, and nationality-based risk profiling, and emphasises strict data protection safeguards, particularly for biometric data. It urges Member States to carry out human rights impact assessments before deploying AI tools and to align their practices with international human rights standards, including the ECHR and the Refugee Convention.