On shaping the EU's external action to address the human rights challenges posed by artificial intelligence

–having regard to the Universal Declaration of Human Rights and other relevant UN human rights treaties and instruments,

–having regard to the International Covenant on Civil and Political Rights of 16 December 1966, and the International Covenant on Economic, Social and Cultural Rights of 16 December 1966,

–having regard to the UN Convention on the Elimination of All Forms of Discrimination against Women of 18 December 1979,

–having regard to the UN Convention on the Rights of Persons with Disabilities of 12 December 2006 and the Optional Protocol thereto, adopted on 13 December 2006,

–having regard to the UN Declaration on the Rights of Indigenous Peoples, adopted by the UN General Assembly on 13 September 2007,

–having regard to the UN Declaration on the Elimination of All Forms of Intolerance and of Discrimination Based on Religion or Belief of 25 November 1981,

–having regard to the UN Declaration on the Rights of Persons Belonging to National or Ethnic, Religious and Linguistic Minorities of 18 December 1992,

–having regard to the UN Convention on the Rights of the Child of 20 November 1989 and the three Optional Protocols thereto,

–having regard to the United Nations’ Global Compact,

–having regard to the Pact for the Future and the Global Digital Compact, adopted at the Summit of the Future in September 2024,

–having regard to UN General Assembly Resolution 78/265 of 1 April 2024 entitled ‘Seizing the opportunities of safe, secure and trustworthy artificial intelligence systems for sustainable development’,

–having regard to the UN General Assembly Resolution 79/325 of 26 August 2025 entitled ‘Terms of reference and modalities for the establishment and functioning of the Independent International Scientific Panel on Artificial Intelligence and the Global Dialogue on Artificial Intelligence Governance’,

–having regard to the UNESCO Recommendation on the Ethics of Artificial Intelligence, adopted on 23 November 2021,

–having regard to the final report of the UN High-Level Advisory Body on Artificial Intelligence of September 2024 entitled ‘Governing AI for Humanity’,

–having regard to the recommendation of the Council of the Organisation for Economic Co-operation and Development of 22 May 2019 on artificial intelligence,

–having regard to the Hiroshima Process International Guiding Principles for Organizations Developing Advanced AI Systems and the Hiroshima Process International Code of Conduct for Organizations Developing Advanced AI Systems, as well as the G7 Leaders’ Statement on AI for Prosperity of 17 June 2025,

–having regard to the G20 AI Principles and Guidelines,

–having regard to the Bletchley Declaration, adopted by countries attending the AI Safety Summit of 1-2 November 2023,

–having regard to the Ministerial Declaration adopted on 4 December 2024 at the Summit of the Global Partnership on Artificial Intelligence,

–having regard to Rule 55 of its Rules of Procedure,

–having regard to the report of the Committee on Foreign Affairs (A10-0000/2026),

A.whereas AI is a global phenomenon that is rapidly being adopted across societies; whereas it unlocks unprecedented opportunities for prosperity, innovation and progress, including the potential to advance the protection and promotion of human rights, democracy and the rule of law in a variety of areas;

B.whereas AI systems need to fundamentally comply with human rights; whereas if inherently designed maliciously, when misused for repressive purposes, or when deployed without adequate safeguards, AI systems may allow for or facilitate human rights violations and may undermine human dignity, human rights, democracy and the rule of law;

C.whereas risks to and challenges for fundamental rights and freedoms arising from AI systems include, but are not limited to, risks to human dignity, data privacy, rights of the child, the right to labour, rights of migrants, environmental rights, democracy and the rule of law and risks of algorithmic discrimination, mass surveillance and suppression of dissent, gender-based violence, an increase in transnational repression, and human rights violations in the deployment of AI systems in automated weapon systems in conflicts and in dual-use equipment;

D.whereas AI also presents significant opportunities for advancing human rights globally, including the monitoring of human rights violations, facilitating access to justice and to information in health and education, detecting disinformation and hate speech, improving delivery of humanitarian and emergency assistance, strengthening transparency in governance and democratic participation, contributing to the UN’s Sustainable Development Goals and moderating the negative impact of climate change;

E.whereas the rapid advancement of AI technologies is profoundly reshaping the global order; whereas the EU must ensure that its external action in AI is guided by a coherent strategic vision, anchored in its core values of democracy, the rule of law and respect for fundamental rights;

F.whereas the EU is championing an alternative, trustworthy model for AI design, development and governance that combines openness, transparency and fundamental rights protection with resilience and innovation, protecting its citizens and offering a level-playing field for companies;

G.whereas strengthening AI partnerships with non-EU countries, particularly candidate countries, countries in the EU neighbourhood and countries in the Global South, should be a priority for EU external action;

General considerations

1.Stresses that human rights, particularly human dignity, equality and non-discrimination, must remain at the centre of AI; stresses that legal human rights obligations apply in all circumstances, and that AI systems need to offer transparency and oversight as a basis for accountability and responsibility, ensuring that remedies are accessible and readily available at every phase of the AI life cycle;

2.Underlines that states have a duty under international law to refrain from and address any unjustified interference with human rights, and to ensure their effective realisation and protection; emphasises that this duty is particularly important in the context of the governance of AI systems;

Risks of AI to human rights

3.Emphasises that data is a crucial component of AI technologies and that safeguarding data privacy is essential for protecting digital human rights;

4.Underlines the risk of AI exacerbating existing social inequalities, particularly for already marginalised groups, as systems may be built on algorithms and/or data that reproduce, perpetuate and exacerbate discriminatory biases;

5.Is deeply concerned that AI has enabled surveillance to become more sophisticated, pervasive and widespread, allowing for the monitoring of entire groups and societies and posing a grave threat to democracy and human rights;

6.Warns that AI-assisted surveillance, even when initially used for legitimate purposes such as law enforcement and security, can easily be repurposed to suppress dissent and target human rights defenders;

7.Strongly condemns the use of AI to perpetuate gender-based violence and undermine gender rights, as seen in the Grok case, which has shown how AI-powered tools can be used to create and spread non-consensual, sexualised content, including deepfakes and other forms of image and video manipulation;

8.Is deeply concerned about the potential harm that AI can cause to children and adolescents;

9.Recognises the impact of AI systems on reshaping global work, labour markets and productivity; emphasises that the effects of AI systems will be profound and long-lasting; warns that the deployment of AI systems cannot be used to justify undermining workers’ rights, dignity and well-being, displacing workers or augmenting skill gaps or deskilling;

10.Notes a growing reliance on AI-powered technologies to monitor borders and migration routes, including along the external borders of the EU;

11.Is deeply concerned about the severe environmental and social implications of AI systems;

12.Is deeply concerned about the growing risks of AI being used to carry out transnational repression;

13.Stresses the crucial role of businesses in ensuring that the development and deployment of AI and data usage respect human rights;

14.Insists that states have a duty to effectively promote and protect human rights against abuses by third parties, including businesses;

15.Calls for the EU institutions and the Member States to lead the efforts in relevant international forums and among international partners to develop AI-specific guidance on corporate responsibility for human rights;

16.Expresses concern about the growing evidence of AI systems being used in warfare;

17.Underlines the importance of applying high human right standards to dual-use items and technologies powered by AI;

Risks of AI to democracy and the rule of law

18.Notes that AI systems have a direct impact on electoral and democratic processes;

19.Emphasises that the deployment of AI systems in electoral processes must comply with the highest standards of transparency, explainability, accountability and independent evaluation;

20.Notes with concern that the use of AI in malicious ways increases the capacities of foreign information manipulation and interference activities, including the potential to undermine the integrity, independence and effectiveness of institutions and processes, as well as the potential to affect information quality;

21.Emphasises that freedom of the press is a fundamental element of democracy; expresses concern about the risks to the diversity, pluralism, inclusiveness, reliability and accountability of the media posed by the concentration of power in the hands of AI providers;

Accountability for AI

22.Notes that accountability for AI-related human rights violations arises at the point in the chain where the system is exploited or used to commit the violation, unless the system was maliciously designed from the outset; emphasises that this approach does not absolve developers of their responsibility;

23.Calls for AI systems to be compliant with the right of victims of human rights violations to an effective remedy;

Recommendations regarding the EU strategy for external action in the field of AI

24.Emphasises the need for the EU to be a leading player in promoting trustworthy AI globally, shaping the global AI landscape and promoting a human rights-based approach to AI development and deployment; calls for the immediate development of a holistic EU strategy for external action in the field of AI to achieve this goal;

25.Calls on the Commission to work towards establishing global minimum standards of protection against serious risks associated with AI;

26.Considers that AI offers a genuine opportunity to address, together with non-EU countries, particular challenges to human rights, such as children’s rights, women’s rights and gender equality, workers’ rights and environmental rights;

27.Calls for the EU and its Member States to support the establishment of participatory governance mechanisms to shape policy responses and AI governance frameworks at international level;

28.Stresses the need to strengthen the EU AI Office and the European External Action Service (EEAS) through the provision of adequate financial and human resources;

29.Calls for the full use of all the instruments at the EU’s disposal to tackle the impact of AI on human rights, including by strengthening the role of the EU Special Representative on Human Rights;

30.Emphasises the need for further discussion on the best ways to introduce the topic of AI in existing EU regulations to ensure that the internal AI policies and rules of the EU strengthen respect for human rights outside the EU;

31.Calls for the Commission and the EEAS to integrate an AI dimension into EU democracy promotion and electoral observation;

32.Stresses the urgent need to develop a strong and resilient EU AI ecosystem that reduces strategic dependencies and strengthens the EU’s strategic autonomy;

33.Recalls the EU’s commitment to a rules-based multilateral order; calls for the EU and its Member States to coordinate their positions in multilateral forums and build partnerships with like-minded partners to promote and advance a shared vision for trustworthy AI;

34.Calls for the EU and its Member States to support and strengthen the UN System in its efforts to promote trustworthy AI;

35.Calls for the EU and its Member States to continue collaborating with the Council of Europe in developing international standards for trustworthy AI;

36.Stresses the need for the EU to support increasing the availability of open-source AI systems and interoperable technical standards, particularly with candidate countries, countries in the EU neighbourhood and countries in the Global South;

37.Calls for the EU to make its financial assistance for AI development and deployment in non-EU countries, including under the upcoming Global Europe instrument and under Global Gateway, conditional on the development and use of trustworthy and rights-respecting AI; stresses that the EU should only finance AI models that uphold human rights standards;

38.Stresses the importance of AI systems used within companies, as they can support companies to comply with key EU laws, such as the Corporate Sustainability Due Diligence Directive3 and the Corporate Sustainability Reporting Directive4, as well as laws against forced labour, in order to ensure that AI supply chains are responsible, sustainable and respectful of human rights and the environment;

39.Stresses that a trustworthy AI ecosystem is a key driver of competitiveness; emphasises that investing in AI systems that respect fundamental human rights is essential for building trust with consumers and international partners, fostering long-term innovation, and creating a sustainable competitive advantage in the global market;

40.Emphasises that supporting AI literacy is of the utmost importance;

Recommendations regarding weapons

41.Supports the UN’s efforts for a legally binding agreement to globally regulate and ban lethal autonomous weapon systems; calls for the EU and its Member States to support such objectives;

42.Condemns the potential human rights violations committed in armed conflicts by means of autonomous AI systems; calls for the EU to increase its support for the International Court of Justice and the International Criminal Court in investigating and prosecuting grave human rights violations committed with the use of AI;

Recommendations regarding AI-assisted mass surveillance

43.Calls for an export ban of AI-assisted mass surveillance systems from the EU to countries where there is a serious risk of these systems being used to violate human rights;

Recommendations regarding dual-use goods

44.Calls for the integration of the principles of the AI Act into the Dual-Use Regulation5, which should lead to the prohibition of exports of AI systems banned by the AI Act and obligate Member States to ensure that all high-risk dual-use AI systems intended for export adhere to the obligations under the AI Act as a condition for export approval;

Recommendations regarding trade

45.Calls for the expansion of the scope of the AI Act to cover exports of AI systems outside the EU for the two main risk categories listed in the AI Act;

46.Calls for the AI provisions in EU trade agreements to promote open-source, trustworthy AI; calls on the Commission to include the promotion of trustworthy AI systems under Global Gateway initiatives;

Recommendations regarding AI opportunities for the promotion of human rights

47.Underscores the potential of AI systems to advance and strengthen human rights and address major challenges at a global level; welcomes efforts by the UN and the Council of Europe to advance this use of AI by means of international governance;

48.Calls on the Commission and the EEAS to support the application of trustworthy AI systems in monitoring human rights violations and supporting access to justice, as well as in the fields of education, healthcare, climate change and energy, and generally in the advancement of the UN’s Sustainable Development Goals;

°

° °

49.Instructs its President to forward this resolution to the Council, the Commission and the Member States.