On 12 November 2026 the European Parliament adopted, by 405 votes to 133 with 60 abstentions, a resolution on the institutional aspects of artificial intelligence in the context of European integration. It asks for parliamentary scrutiny of how AI affects democratic processes, institutional balance and fundamental rights. It is an own-initiative resolution, so it is a political position and does not create legal obligations.
What Parliament adopted and how it voted
Vote records published by HowTheyVote list the text as report A10-0280/2026 (procedure 2025/2118(INI)), voted on 12 November 2026: 405 in favour, 133 against and 60 abstentions. The EPP, S&D, Renew and Greens/EFA groups voted largely in favour, while ECR, Patriots for Europe and Europe of Sovereign Nations members mostly voted against. The Left was split. The Sofia Globe reports the same date and vote count.
The rapporteur was Manolis Kefalogiannis (EPP, Greece). The Constitutional Affairs Committee had backed the report by 21 votes to 3 with no abstentions, according to EU Perspectives, which described the report as expected in plenary during the 19 to 22 October session. The final vote came on 12 November.
What the resolution calls for
According to the Sofia Globe and EU Perspectives, the main calls are these:
- Human control in EU institutions. AI used within the EU institutions should stay human-centred, under human oversight and aligned with EU values. It should not replace human political responsibility or lead to purely automated political decisions.
- Electoral law. MEPs call for reform of European electoral law in light of AI’s influence on political communication and elections. EU Perspectives describes the committee text as proposing stronger action against deepfakes, microtargeting and voter manipulation.
- Charter of Fundamental Rights. The text asks for new rights, including protection of personal identity and likeness and algorithmic transparency.
- A European Parliament AI observatory. A cost-neutral unit would support parliamentary committees in evaluating AI-related legislation and analysing AI’s effect on institutional balance and democracy. EU Perspectives says it would have no regulatory or enforcement powers.
- International rules. Globally significant AI technologies should be governed by an international agreement modelled on nuclear non-proliferation, with the EU taking a leading role.
- European AI capacity. The EU should favour European AI tools, particularly within its institutions, and build strategic autonomy across the AI value chain, while developing safeguards without unduly slowing innovation.
What a non-binding resolution can and cannot do
An own-initiative resolution states Parliament’s position and invites the Commission and Member States to act. It does not amend the EU AI Act or any other law. Changing the Charter or electoral law would need separate legislative or treaty procedures involving other institutions, so the practical effect depends on what the Commission proposes next. The resolution sits alongside existing instruments such as the Digital Services Act, which already governs systemic risks on very large platforms.
What it means for researchers and research offices
Published accounts of the text do not describe provisions on research funding, researcher data access or disinformation research specifically, so universities should not read the resolution as creating new rights or duties for research teams. Its relevance is indirect and, in places, editorial interpretation:
- Teams studying elections, political advertising or AI-generated disinformation can treat the electoral-law and deepfake language as a signal of where EU policy attention is heading. The Commission’s parallel consultation on draft DSA election guidelines is the more concrete near-term vehicle.
- The proposed AI observatory could become a new source of institutional demand for analysis of AI’s effect on democratic processes. No budget, mandate or call for expertise has been announced.
- Research offices tracking AI regulation should keep this resolution separate from binding obligations under the EU AI Act’s high-risk rules.
What to watch next
Watch for a Commission response, any proposal touching electoral rules or the Charter, and whether Parliament sets up the observatory. Until then, the resolution is best read as a statement of Parliament’s priorities rather than a change in the rules that govern research.
