ChatGPT and other AI tools have long been part of the dissertation writing process. But in 2026, the rules changed dramatically for Ukrainian master’s and doctoral students: Ukraine’s first law explicitly regulating the use of AI in academic activities was enacted. And ignorance of these rules is no longer an excuse.
Let's break it down: what's allowed now, what's prohibited, how inspections are conducted, and what the penalties are for violations.
Law No. 4742-IX: What It Says About AI
On December 18, 2025, the Verkhovna Rada passed the Law “On Academic Integrity”—with 260 votes in favor.
It took effect on February 1, 2026. Full implementation begins on July 31, 2026: by that date, each university is required to adopt its own internal regulations.
This is the first law in Ukraine—and one of the first in Europe—to regulate the use of AI in scientific research at the legislative level.
What does the law define as “unethical use of AI”?
The law introduces a new concept: “unethical use of artificial intelligence outputs”—and classifies it as a separate type of academic misconduct, on par with plagiarism and data fabrication.
Key indicator of a violation: presenting AI-generated text as one’s own without disclosing it. If an AI generated a text excerpt or analysis, this must be noted, specifying the tool and the method used.
An important clarification: Using AI to search for information (with subsequent verification by the author) or to edit one’s own text is NOT a violation. A violation occurs only when AI-generated text is presented as one’s own without attribution (Position of the Ukrainian Student League, which participated in drafting the law.)
Three responsibilities when using AI in a dissertation:
- Transparency — specify which tools were used (ChatGPT, Claude, Gemini, etc.), for which tasks, and at which stage;
- Labeling — AI-generated text must be labeled in the paper with a reference to the program;
- Personal Responsibility — All sections written using AI must be reviewed and edited by the author personally (requirement Cabinet of Ministers of Ukraine Resolution No. 928).
What are the penalties for violations?
For graduate students / Ph.D. candidates:
Each university determines the specific sanctions for “unethical use of AI” independently in its internal regulations. Penalties range from a warning and having to redo the work to academic dismissal. Plagiarism in a defended dissertation results in the automatic revocation of the degree award.
For doctoral students / candidates for a Doctor of Science degree:
It's tougher here. Cabinet of Ministers Resolution No. 928 Effective July 30, 2025, the law explicitly states: if a dissertation is found to have been “written by AI” without disclosure, the doctoral degree will be revoked. This is the first provision in Ukraine to designate AI as a direct grounds for revoking an academic degree.
In addition, for anyone who has violated the rules: a ban on participating in grants and mobility programs, inclusion in the Register of Violators, and—in the case of purchasing a dissertation—a 10-year ban on holding leadership positions at higher education institutions.
The law does not apply retroactively to AI-related violations: works protected as of February 1, 2026, are not subject to review for “unethical use of AI” under the new regulations. However, plagiarism is subject to review regardless of the date of defense (Supreme Court precedent dated June 3, 2025).
How They Evaluate: Tools Used by Ukrainian Universities
StrikePlagiarism — actual standard for 2025–2026
This tool is currently the most widely used among Ukrainian universities. The National Repository of Academic Texts (UkrINTEI) officially recommends it as the primary tool for verification.
Starting in December 2025 StrikePlagiarism It features a comprehensive AI content detection module: it analyzes stylistic and linguistic patterns and generates an official PDF report that universities accept. It detects content generated by ChatGPT and other large language models (LLMs) in more than 100 languages, including Ukrainian.
Specific universities with contracts for StrikePlagiarism: DUET University (agreement dated August 28, 2024), Gonchar Dnipro National University, Boris Grinchenko Kyiv Municipal University, and others.
Here's what the verification process looks like in practice:
- The author submits the dissertation through the higher education institution's system or to the department;
- The person in charge uploads the file to StrikePlagiarism;
- The system generates two reports: similarity to other texts + percentage of text likely to be machine-generated;
- If the scores exceed the university's threshold, the case is referred to the Academic Integrity Committee.
Important: No detector is 100% reliable. Manufacturers StrikePlagiarism and Turnitin They officially acknowledge that the report provides a “probability ratio” rather than a verdict. A polygraph result cannot serve as the sole evidence in disciplinary proceedings.
What the Ministry of Education and Science and the Ministry of Digital Transformation Say: Where Is the Line Between “Allowed” and “Not Allowed”?
In April 2025, the Ministry of Education and Science and the Ministry of Digital Transformation jointly published the “Recommendations on the Responsible Use of AI in Higher Education Institutions.” The position is clear: the Ministry of Education and Science does not ban AI, but requires its responsible and transparent use.
What is allowed:
- Use AI to search for and organize information—with independent verification of each source;
- Ask AI to explain complex concepts or methodological approaches;
- Receive feedback from AI on your own text: grammar, logic, and structure;
- Use AI to translate or adapt quotes from foreign sources;
- Formulate research hypotheses—followed by one's own scientific analysis.
What is prohibited:
- Submit AI-generated text as your own without disclosing its source;
- Generate the sections of the dissertation (literature review, methodology, analysis) and insert them "as is";
- To use AI to generate scientific conclusions without verifying them independently;
- Citing AI-generated sources without verifying their existence carries the risk of “hallucinations.”
The Real Numbers: What's Already Happening in Ukraine
According to data from the National Agency for Higher Education Quality Assurance (NAZYAO) and the Ministry of Education and Science, in 2025, the following were recorded: 50 complaints regarding academic misconduct, 10 dissertations found to contain plagiarism, and 8 individuals stripped of their academic degrees.
Important context: All 8 cases of degree revocation in 2025 were due to plagiarism, not AI-generated content. The law recognizing AI as a standalone violation did not take effect until February 1, 2026, so there are currently no specific cases of “degree revocation due to AI” in public records. However, the regulatory framework is already in place.
In December 2023, Igor Sikorsky Kyiv Polytechnic Institute adopted a “Policy on the Use of AI in Academic Activities”—the first among major public universities to do so. In June 2025, the university established a separate Center for AI Application Strategies.
A Practical Guide: How to Protect Yourself
1. Keep an AI journal starting from the first day of work on your dissertation
Make a note of: which tool you used, when, for what purpose, and what query you entered. This will serve as the basis for your declaration and protect you in the event of a possible appeal.
2. Include the declaration in the body of the dissertation
In accordance with Law No. 4742-IX and Cabinet of Ministers Resolution No. 928, the wording may be as follows:
“The following artificial intelligence tools were used in this work: [names]. They were used for [specific tasks: literature search, grammar checking, etc.]. All sections written with the use of AI have been reviewed and edited by the author personally. The scientific arguments, conclusions, and results are the author’s own intellectual contribution.”
3. Check your work yourself before submitting it
Upload the text to StrikePlagiarism (through the university) or free online tools. If a passage has a high AI score, rephrase it in your own words. Declared AI-generated content is considered fair use; hidden AI-generated content is considered a violation.
4. Check each source manually
If you used AI to search for literature, verify each link in Google Scholar or a library. AI “hallucinations” (non-existent citations) are one of the easiest signs to spot and among the most dangerous for the author.
5. Clarify your university's position
By July 31, 2026, every higher education institution must adopt internal regulations regarding AI. Ask the graduate studies office: what threshold for AI content is considered critical, what form the declaration takes at your institution, and what tool is used to check dissertations.
Conclusion
AI in dissertations is no longer a matter for the future. It is a reality that has been legally formalized. The law has taken effect, the Cabinet of Ministers has adopted a resolution, and universities are developing internal regulations. The rules of the game have changed—and it’s better to know them in advance than to find out during your defense.
The good news is: it’s okay to use AI. And it’s perfectly fine to search for references, check grammar, and generate hypotheses. The main thing is not to hide it. Documented and disclosed AI use is not a violation. Hidden use is.
By July 31, 2026, every university is required to adopt its own rules. If your institution hasn’t done so yet, check now—don’t wait. And be sure to document in your dissertation exactly how you used AI. A single paragraph in your declaration could protect years of work.
We can help you review your work and prepare it for submission, taking into account the current requirements regarding academic integrity and the use of AI.
What we can do:
Let's start with free consultation: We'll analyze your situation, determine what needs to be reviewed or refined, and propose the best course of action.
