Policy explainerAll disciplines

The EU AI Act: what it means for doctoral researchers

The EU Artificial Intelligence Act regulates AI by risk, but it largely excludes AI built purely for scientific research. Here is where that exclusion ends, which dates matter after the 2026 Digital Omnibus, and what a PhD student should check (European Parliament & Council of the European Union, 2024; European Parliament & Council of the European Union, 2026).

PhD Research Lab InsightsEditorial team, reviewed by a subject specialist
Posted
3 October 2026
Last checked
3 October 2026
Reading time
7 min
Sources
8
Read the official text

At a glance

Issued by
European Parliament and Council of the European Union
Instrument
Regulation (EU) 2024/1689 (Artificial Intelligence Act), directly applicable in every Member State
Applies to
Providers and deployers of AI systems, providers of general-purpose AI models, importers and distributors, including those outside the EU whose AI output is used in the EU
Research
Excluded when AI is developed and put into service solely for scientific research and development (Article 2(6) and 2(8))
Status
In force since 1 August 2024; applies in phases; amended by Regulation (EU) 2026/1744 (Digital Omnibus on AI)
Read the official text ↗

Timeline

  1. 12 July 2024Published in the Official Journal of the European Union as Regulation (EU) 2024/1689 (European Parliament & Council of the European Union, 2024).
  2. 1 August 2024Enters into force, twenty days after publication (European Commission, n.d.-c).
  3. 2 February 2025General provisions, including AI literacy (Article 4), and the prohibited practices (Article 5) apply (European Parliament & Council of the European Union, 2024).
  4. 2 August 2025Obligations for general-purpose AI models, governance and penalties apply (European Parliament & Council of the European Union, 2024).
  5. 27 July 2026Digital Omnibus on AI, Regulation (EU) 2026/1744, enters into force after publication on 24 July 2026 (European Parliament & Council of the European Union, 2026).
  6. 2 August 2026Most remaining provisions apply, including the Article 50 transparency obligations (European Commission, n.d.-c).
  7. 2 December 2026New prohibitions on AI generating non-consensual intimate imagery and child sexual abuse material apply; deadline for marking AI-generated content from systems placed on the market before 2 August 2026 (European Commission, n.d.-c; Council of the European Union, 2026).
  8. 2 December 2027Rules for stand-alone high-risk AI systems listed in Annex III apply (originally 2 August 2026) (European Parliament & Council of the European Union, 2026; European Commission, n.d.-c).
  9. 2 August 2028Rules for high-risk AI embedded in products covered by Annex I legislation apply (originally 2 August 2027) (European Parliament & Council of the European Union, 2026; European Commission, n.d.-c).

What it requires

Your checklist

  • Write down, in your data management plan, whether your AI system is built and used solely for scientific research and development, and how you will keep it that way.
  • Note any point where the system could leave the lab: a spin-out, a public demo, a tool given to a partner hospital or school, or a pilot with real users. Those steps may move it into the Act's scope (Meszaros et al., 2026).
  • If you plan testing in real-world conditions, ask your institution early: Article 2(8) states that such testing is not covered by the research exclusion (European Commission, n.d.-a).
  • If you build or publish a chatbot or generative tool, plan for the Article 50 duties to inform users and to mark synthetic output (European Commission, n.d.-b).
  • Keep your GDPR work separate and complete: lawful basis, data protection impact assessment where needed, and ethics approval (European Parliament & Council of the European Union, 2016).
  • Check whether your university has AI-literacy training or an AI use policy, and record that you completed it.

What the AI Act is and how it works

The Artificial Intelligence Act is an EU regulation, so it applies directly in every Member State without national transposition (European Parliament & Council of the European Union, 2024). It does not regulate "AI" as a technology. It regulates AI systems and general-purpose AI models according to the risk their use poses to health, safety and fundamental rights.

The structure has four layers. A small number of practices are prohibited. High-risk systems, defined through Annex I (products already covered by EU safety legislation) and Annex III (listed use areas such as education, employment and access to essential services), carry the heaviest obligations. Some systems carry only transparency duties under Article 50. Everything else is minimal risk and largely unregulated by the Act, although other law still applies (European Parliament & Council of the European Union, 2024).

Separately, providers of general-purpose AI models, the large models that underpin many chat and coding tools, have their own obligations under Chapter V, which have applied since 2 August 2025 (European Commission, n.d.-c).

The research exclusion: what Article 2(6) and 2(8) actually say

Two provisions matter most for doctoral researchers. Article 2(6) states that the Regulation "does not apply to AI systems or AI models, including their output, specifically developed and put into service for the sole purpose of scientific research and development" (European Commission, n.d.-a).

Article 2(8) states that the Regulation "does not apply to any research, testing or development activity regarding AI systems or AI models prior to their being placed on the market or put into service". It adds that such activities must comply with applicable Union law and that testing in real-world conditions is not covered by the exclusion (European Commission, n.d.-a).

Commentators have pointed out that these boundaries are hard to apply to modern AI research, where academic and commercial work overlap, models are reused across projects and "real-world testing" is not sharply defined (Meszaros et al., 2026). Treat the exclusion as a condition you need to document, not a blanket exemption for anything done at a university.

When you use AI tools rather than build them

Most PhD students are deployers or simply users of AI tools, not providers. Using a general-purpose chatbot to summarise papers or debug code does not, on its own, make you subject to high-risk obligations. Your obligations come mainly from your university, your funder, your publisher and data protection law.

Two parts of the Act still touch everyday research. Article 50(4) requires deployers who publish deepfakes to disclose that the content is artificially generated or manipulated, with lighter rules for evidently artistic or creative work, and requires disclosure of AI-generated text published to inform the public unless it has undergone human review or editorial control and someone holds editorial responsibility (European Commission, n.d.-b). If your project uses emotion recognition or biometric categorisation on participants, Article 50(3) requires you to inform the people exposed to it (European Commission, n.d.-b).

GDPR, and what changed in 2026

The AI Act sits alongside the GDPR rather than replacing it. Article 2(7) confirms that Union law on personal data, privacy and confidentiality of communications applies to personal data processed in connection with the Act (European Commission, n.d.-a). A research exclusion under the AI Act therefore says nothing about whether your processing of participant data is lawful; that question is answered under the GDPR and national research rules (European Parliament & Council of the European Union, 2016).

In July 2026 the EU adopted the Digital Omnibus on AI, Regulation (EU) 2026/1744, published on 24 July and in force from 27 July 2026 (European Parliament & Council of the European Union, 2026). It postponed the high-risk rules to 2 December 2027 for Annex III systems and 2 August 2028 for systems embedded in Annex I products, set 2 December 2026 as the marking deadline for generative systems already on the market, and added new prohibitions on AI that generates non-consensual intimate imagery or child sexual abuse material (Council of the European Union, 2026; European Commission, n.d.-c). It did not remove the research exclusion. Dates may still be refined by guidance, so check the official timeline before relying on a deadline.

What to do now

  • Read Article 2 of the Act alongside your data management plan and note which exclusion you rely on (European Commission, n.d.-a).
  • Bookmark the Commission's implementation timeline and re-check it before any deadline-dependent decision (European Commission, n.d.-c).
  • Raise any plan to share, deploy or field-test your AI system with your supervisor and research ethics committee before it happens.
Your next step

Does this rule affect your project?

  1. 1Run through the checklist above
  2. 2Ask a mentor to check your plan, data or paper
  3. 3Keep a record of what you did

Questions scholars ask

Does the AI Act apply to my thesis project?

If your AI system or model is developed and put into service solely for scientific research and development, Article 2(6) excludes it. The position changes if it is placed on the market, put into service for other purposes, or tested in real-world conditions (European Commission, n.d.-a).

I only use ChatGPT-style tools. Do I have obligations under the Act?

Ordinary use for reading, drafting or coding does not make you a provider of a high-risk system. Your university, funder and publisher policies on disclosing AI use still apply, and Article 50 disclosure rules can apply if you publish deepfakes or AI-generated public-information text without editorial review (European Commission, n.d.-b).

Does the research exclusion mean I can skip GDPR steps?

No. Article 2(7) keeps EU data protection law fully in place, so lawful basis, data minimisation, impact assessments and ethics approval are still required where personal data are involved (European Commission, n.d.-a; European Parliament & Council of the European Union, 2016).

Were the deadlines delayed?

Yes, in part. Regulation (EU) 2026/1744 moved the high-risk dates to 2 December 2027 (Annex III) and 2 August 2028 (Annex I). Most other provisions, including the Article 50 transparency duties, apply from 2 August 2026 (European Parliament & Council of the European Union, 2026; European Commission, n.d.-c).

I am outside the EU. Can the Act still matter to me?

It can, if you place an AI system on the EU market or its output is used in the EU, or if you collaborate with EU institutions that do. The research exclusion applies in the same way (European Parliament & Council of the European Union, 2024).

Sources

  1. Council of the European Union. (2026, June 29). Artificial intelligence: Council gives final green light to simplify and streamline rules [Press release]. https://www.consilium.europa.eu/en/press/press-releases/2026/06/29/artificial-intelligence-council-gives-final-green-light-to-simplify-and-streamline-rules/
  2. European Commission. (n.d.-a). Article 2: Scope. AI Act Service Desk. Retrieved 3 October 2026. https://ai-act-service-desk.ec.europa.eu/en/ai-act/article-2
  3. European Commission. (n.d.-b). Article 50: Transparency obligations for providers and deployers of certain AI systems. AI Act Service Desk. Retrieved 3 October 2026. https://ai-act-service-desk.ec.europa.eu/en/ai-act/article-50
  4. European Commission. (n.d.-c). Timeline for the implementation of the EU AI Act. AI Act Service Desk. Retrieved 3 October 2026. https://ai-act-service-desk.ec.europa.eu/en/ai-act/timeline/timeline-implementation-eu-ai-act
  5. European Parliament & Council of the European Union. (2016). Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data (General Data Protection Regulation). Official Journal of the European Union, L 119, 1–88. https://eur-lex.europa.eu/eli/reg/2016/679/oj
  6. European Parliament & Council of the European Union. (2024). Regulation (EU) 2024/1689 of the European Parliament and of the Council of 13 June 2024 laying down harmonised rules on artificial intelligence (Artificial Intelligence Act). Official Journal of the European Union, L, 2024/1689. https://eur-lex.europa.eu/eli/reg/2024/1689/oj
  7. European Parliament & Council of the European Union. (2026). Regulation (EU) 2026/1744 of the European Parliament and of the Council of 8 July 2026 amending Regulations (EU) 2024/1689, (EU) 2018/1139 and (EU) 2023/1230 as regards the simplification of the implementation of harmonised rules on artificial intelligence (Digital Omnibus on AI). Official Journal of the European Union, L, 2026/1744. https://eur-lex.europa.eu/eli/reg/2026/1744/oj/eng
  8. Meszaros, J., Huys, I., & Ioannidis, J. P. A. (2026). Challenges in applying the EU AI act research exemptions to contemporary AI research. npj Digital Medicine, 9, 288. https://doi.org/10.1038/s41746-025-02263-0

How to cite this page

Choose your referencing style, then copy.

PhD Research Lab Insights. (2026, October 3). The EU AI Act: what it means for doctoral researchers. PhD Research Lab. https://www.phdassistance.com/insights/eu-ai-act-for-doctoral-researchers/

PhD Research Lab Insights. “The EU AI Act: What It Means for Doctoral Researchers.” PhD Research Lab, 3 Oct. 2026, www.phdassistance.com/insights/eu-ai-act-for-doctoral-researchers. Accessed [access date].

PhD Research Lab Insights (2026) The EU AI Act: what it means for doctoral researchers. PhD Research Lab. Available at: https://www.phdassistance.com/insights/eu-ai-act-for-doctoral-researchers/ (Accessed: [access date]).

PhD Research Lab Insights. “The EU AI Act: What It Means for Doctoral Researchers.” PhD Research Lab. October 3, 2026. https://www.phdassistance.com/insights/eu-ai-act-for-doctoral-researchers.

PhD Research Lab Insights, ‘The EU AI Act: What It Means for Doctoral Researchers’ (PhD Research Lab, 3 October 2026) <https://www.phdassistance.com/insights/eu-ai-act-for-doctoral-researchers> accessed [access date].

PhD Research Lab Insights. The EU AI Act: what it means for doctoral researchers [Internet]. PhD Research Lab; 2026 Oct 3 [cited [access date]]. Available from: https://www.phdassistance.com/insights/eu-ai-act-for-doctoral-researchers/

PhD Research Lab Insights, “The EU AI Act: what it means for doctoral researchers,” PhD Research Lab, Oct. 3, 2026. [Online]. Available: https://www.phdassistance.com/insights/eu-ai-act-for-doctoral-researchers. [Accessed: [access date]].

@online{eu_ai_act_for_doctoral_researchers_2026,
  author = {{PhD Research Lab Insights}},
  title = {{The EU AI Act: what it means for doctoral researchers}},
  organization = {PhD Research Lab},
  date = {2026-10-03},
  url = {https://www.phdassistance.com/insights/eu-ai-act-for-doctoral-researchers},
  urldate = {[access date]}
}

  • AI Act
  • EU regulation
  • research exemption
  • GDPR
  • research ethics
The EU AI Act: what it means for doctoral researchersRegulation explainer
Official text