The issue of academic integrity has gained considerable prominence in the last couple of decades. In this article I will attempt to summarize what I see as some key issues, cite a few authoritative articles, and communicate some impressions formed from personal experience as an academic.¹

I left India for the first time to join the physics program at Stony Brook University, New York. Everything was new and exciting at this time, especially for someone coming from India, and the formative impact of these five years on me cannot be overemphasized. My first PhD guide, Daniel Z. Freedman, imparted a valuable lesson to me on the importance of honesty and transparency in research. While working on my first research project under his guidance, I came to know that a postdoctoral fellow in the same department was working on something closely related. Freedman instructed me to meet this person, tell him unconditionally what I was doing and how far I had reached, and ask him to respond similarly. This went very well and we ended up publishing independent papers using different methods but with overlapping results. I learned the virtue of transparency, though of course I was fortunate that the other party also acted responsibly.

I did not give conscious thought to academic ethics in those days, and it rarely came up as a subject of discussion in my presence beyond the occasional anecdote about some specific scientist. I do not remember any case of paper retraction for misconduct, or even of copying in examinations, during my five years at Stony Brook. This continued to hold true in the initial years following my return to India in 1984 as a postdoctoral fellow and then faculty member at the Tata Institute of Fundamental Research (TIFR).

Explore the full InterDialogue with Sunil Mukhi.

It took the advent of the internet and a very egregious case in India² to focus my attention on academic ethics, the ways in which it could be subverted, and the broad consequences of academic misconduct. It was 2001 and I was on sabbatical at the Institute for Advanced Study, Princeton. Internet had become widespread and easy to use, even though Gmail, Facebook, and YouTube were yet to be invented. One day I received an email from a physicist friend in India telling me that Prof. B. S. Rajput, the vice-chancellor³ of a large Indian university, together with his PhD students, had plagiarized several papers originally published by a distinguished academic in the US. Efforts to publicize the problem seemed to be getting nowhere, and I sensed something had to be done to change this.

Fortunately, the papers were all in my broad field of research—supersymmetry,⁴ quantum field theory, and string theory.⁵ I downloaded copies of these papers, both the original and plagiarized versions, and sat down to inspect them. In collaboration with a few Indian colleagues, I then created a page titled “Physics Plagiarism Alert” using the now-obsolete GeoCities webpage platform (the page is still visible⁶). On it, I posted textual excerpts of this vice-chancellor’s papers side-by-side with the original papers from which they were plagiarized. Most often, the plagiarism involved minor paraphrasing that did not successfully cover up the act. Sometimes, it was even verbatim.

The purpose of the webpage was to present the facts in an easy-to-compare format, so that we could convince academics in the same field to endorse it by confirming that in their view, serious plagiarism had occurred. I hoped this would help to build pressure on authorities to look into the matter. This campaign went quite well: we emailed a few dozen friends and colleagues and most of them viewed the page and emailed us with their endorsement. The list of endorsers grew steadily to around fifty and we updated it regularly at the foot of the webpage for the public to see. Gradually, this issue started to be discussed in the Indian media.⁷ The final push came when a group of Stanford academics wrote about the matter to the President of India, Dr. A. P. J. Abdul Kalam, himself a former aerospace engineer, and asked him to intervene. He constituted an empowered committee to investigate the plagiarism charges. I learned later that as soon as the retired judge heading the committee viewed our webpage, he agreed that serious plagiarism had occurred. The vice-chancellor had to resign.⁸

Explore the full InterDialogue with Sunil Mukhi.

This episode had implications in India. Just a year later the Indian Academy of Sciences set up a standing committee to look into allegations of ethical misconduct. I have been a member of this committee since its inception. Its first assignment was to draft ethical guidelines for the Academy.⁹ To my knowledge these were the first broad-based academic ethics guidelines of an institution—as distinct from specific guidelines for medical ethics,¹⁰ for example—in India.

I do not mean to imply that the vice-chancellor story was the first major case of ethical misconduct that received publicity in India. The Himalayan fossil hoax was perpetrated by a professor from Chandigarh, India during the 1970s and 1980s and exposed by Australian and European scientists. But the case I have described may have been among the earliest to be investigated and exposed by Indian scientists, with the government playing an active and responsible role through the President of India.

The ethics committee of the Academy met a few times during the tenure of its wonderful founding chair, the late Prof. R. Kumar, from whom I learned profound lessons on how to investigate professionally and dispassionately. We even took action on some high-profile cases that I am not at liberty to discuss. But the committee then became moribund for many years. Some subsequent presidents of the Academy were even unaware of its existence, until it was revived by incoming president Ram Ramaswamy in 2015, and a number of misconduct complaints started to come in.

Explore the full InterDialogue with Ram Ramaswamy.

One thing I remember noticing was that, whenever their questionable practices came to light, the perpetrators would inevitably make things worse. Rather than admit an error and try to make it right, they would feign ignorance, pass the blame on to students, hide their correspondence with the journal—in short, everything short of coming clean and fixing the problem. This was particularly annoying in cases where the initial misconduct was minor and would at best have led to a slap on the wrist.

I had hoped that the publicity attaching to the “vice-chancellor case” would kick off a period of introspection among the Indian scientific community regarding the extent of ethical problems within our fold. But that did not happen for a long time. The University Grants Commission of India came out with its “Promotion of Academic Integrity and Prevention of Plagiarism in Higher Educational Institutions”¹¹ only in 2018, and the Indian National Science Academy only implemented actionable ethics guidelines¹² in February 2026. Both are creditable milestones but should have been achieved decades earlier.

Some Takeaways from Actual Cases

Why do people indulge in plagiarism and data fabrication in scientific research? Although my own experience has been primarily with cases in India, the question is of significantly wider interest. An article in The Conversation¹³ argues thus: “The ‘publish or perish’ culture within academia fuels¹⁴ scientific misconduct. It puts pressure on academics to meet publication quotas. It also rewards them for greater research output, in the form of promotions, funding and recognition. And this can mean research quality is sacrificed for quantity.”

This jibes with my own estimation that the most powerful driver of academic misconduct is the desire to be seen as successful and, through that success, to attain promotions, awards, and leadership positions. However, in a few rare cases, misconduct has occurred because a scientist was desperate to fulfill a grant commitment or publish in time for a promotion review.

One should not, of course, include genuine errors in “misconduct.” But the article quoted above finds that “fewer than 20% of all retractions are due to honest mistakes,” and so retractions of papers¹⁵ have come to be seen as a proxy for misconduct.

An important concept that tends not to be properly understood is unintentional plagiarism. A page¹⁶ on the website of Oxford University explains it thus: “Not all cases of plagiarism arise from a deliberate intention to cheat. Sometimes students may omit to take down citation details when taking notes, or they may be genuinely ignorant of referencing conventions.” An article¹⁷ on the website of the International Center for Academic Integrity reviews a research study conducted in Puerto Rico which found that “knowledge about plagiarism among all participating graduate students was poor, with an overall average score of 43%.” 

Personally, I find it surprising that so many references to plagiarism focus exclusively on students. I have encountered numerous researchers, including faculty members and scientific staff, who did not know what it meant, and I will discuss one such case below.

Given that every research paper relies on older ones to some extent, what is the acceptable way to introduce older results? The key idea is to ensure that what one writes will not induce even a casual reader to believe these ideas are original. A limited amount of original material may be re-used in quotes and with a reference to the original author where the quotes start (not merely in a generic place in one’s article). Alternatively, the original work may be summarized in one’s own words and without quotes; for example, “The work of X, Y, and Z [reference] showed that…” The constructions of the original authors should not be re-used, even lightly disguised. The Oxford University page cited above provides a useful example, with five instances that summarize an original source in ways that count as “plagiarized,” and two more that do so in ways that are acceptable. It is imperative for educational institutions to carry out training workshops for staff and students that communicate important subtleties like this one. Unfortunately, this type of careful instruction remains rare in Indian institutions.

Explore the full InterDialogue with Sunil Mukhi.

During an ethics investigation in India, I spoke to an author of a manuscript who had been tasked by co-authors to type a short review of standard material for the introduction section. This author had attached the original paper to the side of their monitor and consulted it regularly while typing the introduction. Inevitably, a degree of plagiarism crept in. For this reason, it is commonly advised not to keep original sources open while attempting to review their contents. One should read the original, put it away, and then put down one’s own understanding of what was done, of course always with due attribution.

Global Perspective

I have spent all my academic life working in India, despite a couple of long sabbaticals and numerous shorter visits abroad. My administrative experience in particular, including my roles in ethics committees, had until recently been entirely limited to institutions within India. Naturally, I would read news reports on ethical misconduct cases from around the world, but these did not provide access to any of the microscopic details that would enable me to appreciate what sort of misconduct occurs outside India and how it is handled.

This changed, to some extent, when The World Academy of Sciences (TWAS) asked me to chair their first ethics committee, with members drawn from around the world. TWAS is a program unit of UNESCO with an office in Italy. Its mission statement establishes its mandate to “recognize, support and promote excellence in scientific research in the developing world,” and most TWAS Fellows hail from developing countries.

Explore the full InterDialogue with Marcelo Knobel.

We first met to formulate a code of conduct, which is now publicly available,¹⁸ and then looked into cases referred to us by the Academy. Confidentiality prevents me from disclosing the nature of the cases we considered and even the distribution of cases by region, though in view of the demographics of TWAS, it is clear that most of these come from developing countries.

I can recount some informal interactions with delegates at a recent TWAS meeting that impacted me. One of them asked me if it was ethically acceptable to provide financial support to rural people to travel to a university for a health survey, as they had to pass through disturbed regions where violent conflicts were ongoing, and it would not be safe to expect them to make their own travel arrangements. This individual was concerned that paying them money might be seen as unethically biasing the survey. I responded that it seemed quite reasonable to reimburse costs, that the safety of human subjects was worth any possible effort one could make, and that this need not be seen as introducing bias. This incident made me aware of contexts and situations very different from those familiar to me, and I was touched to realize that sensitive, ethically minded academics were dealing with them.

I also heard about some of the more familiar problems pertaining to publication misconduct. One of the themes that seemed to come up often was of powerful people misusing their positions to commit ethical misconduct and getting away with it because others were too afraid to complain. This resonated closely with my Indian experience, and I got the impression that the sociology around ethical misconduct in academia is rather similar across the developing world.

This is by no means to say that academia in the developed world does not have its ethical problems, including very serious ones. One of the most horrific cases I learned about involved Magdalena Koziol, a young researcher in biology at Yale University, who discovered that her zebrafish were being poisoned overnight by a rival. A report on this case was published¹⁹ in Science.²⁰ But beyond sabotage, what I am inclined to take away from this case is that it featured deliberate, preplanned misconduct with malicious intent.²¹ 

Ranking Systems

In recent years, ranking systems have come to dominate the discourse on academic success, both for institutions and individuals. Universities are ranked in numerical order by agencies (some private, others public) based on so-called “parameters,” which invariably include the number of publications and the number of citations. This has unfortunately introduced new incentives for ethical misconduct; this article in a leading Indian daily newspaper,²² which I co-authored,²³ highlights the problem.

Unscrupulous universities have devised ingenious unethical ways to boost these parameters and thereby enhance their rankings. In early 2025, Lebanese American academic and bibliometrics expert Lokman Meho published an article titled “Using bibliometrics to detect questionable authorship and affiliation practices and their impact on global research metrics: A case study of 14 universities.”²⁴ The article flags universities whose output increased at a dramatic rate just in the last few years and which in many cases employ “hyperprolific authors.” These two features are thought to be evidence of activity deliberately targeted at an improvement in rankings, without any significant improvement in intellectual contribution to knowledge.

I have come to know of universities that openly admit to incentivizing their undergraduate students to carry out simple surveys (that take no more than a day or two) and present these in the form of “publications.” These amateur surveys are conducted by asking a few dozen peers for their views on literally anything—preferred foods, relationships, management of emotions. The resulting data is used to generate manuscripts generously inflated with tables and charts. These universities then pay publication fees to a newly manufactured journal to publish these essentially content-free manuscripts. The total number of publications per year from the university, and consequently its rank on the publications parameter, now rapidly rises. This in turn leads more students to enroll in that university, which easily recovers its publication expenses. Here we have a perfect closed cycle, where unfortunately no worthwhile knowledge is produced. Citation rankings have been similarly gamed by insisting that everyone who publishes from a given university must cite a dozen other papers (never mind their relevance) from the same university.

In terms of the global prevalence of this problem, again it appears that institutions in the developing world are more likely to indulge in gaming the rankings system. For example, Meho’s article cited above studies publication metrics from specific institutions in Saudi Arabia, Egypt, Lebanon, Iraq, and India, and compares them with a “control group” of US universities, finding that the former have unexpected, sharp rises in their metrics over a short span of four years while the latter do not. A counterargument has been made that developing countries are seeing rapid growth in their academic communities, so these rises may be expected. The issue clearly merits wider and more careful analysis.

Artificial Intelligence

The widespread availability of large language models (LLMs) like ChatGPT has introduced a new and dramatic twist to the ethics story. An LLM can create entire articles complete with figures and even videos, which academics can then pass off as their own work. 

A common reaction nowadays is to feed each manuscript to an LLM and ask if it is likely to be AI-generated. Unfortunately, this becomes a cat-and-mouse game where the same software simultaneously provides the means to cheat on a large scale and is then tasked with detecting whether such cheating has occurred. It is not clear how an internal ethical conflict between an LLM and itself ends. But in this high-stakes game, an academic bent on committing misconduct might use a very advanced LLM (through a paid subscription) while the university’s academic cell, which would have to assess hundreds of manuscripts, might be obliged to use a free version which is less competent at detection. It seems that nowadays a meaningful assessment of a student’s abilities and performance can only come from a one-on-one meeting, which is of course highly impractical in universities with limited staff and a massive enrollment.

The article “LLM Use, Cheating, and Academic Integrity in Software Engineering Education”²⁵ reports on a survey of over a hundred students across several countries. In the survey, students self-reported their own use of LLMs in ways that were disallowed under the guidelines for their courses. As the title indicates, the study was restricted to a specific area of education at the outset, but valuable general lessons emerged. Reasons cited by students for their own misconduct included “Heavy workload or overlapping deadlines,” “Limited oversight or checking by instructors,” “Unclear or missing rules about AI use,” and “Peer norms encouraging similar behavior.” The authors argued for “a need for clearer alignment between assessment design, learning objectives, and expectations for LLM use.” In my own informal assessment, peer norms (coupled with limited oversight) are a very serious problem in India, and I will return to this point below.

Of course, LLMs do not just enable cheating. They are a valuable asset to academic research as they can perform tedious calculations much faster than humans, and also integrate fragmented knowledge due to the enormous data sets on which they have been trained. They have been responsible for results in mathematics and theoretical physics²⁶ that experts consider “new.”²⁷ To a large extent we have to accept this as a development of our era,²⁸ like the internet, television, computers, and even the printing press of earlier eras.

With each jump in technology, the role of humans has been redefined, and today we are not completely clear what scientific research will mean in the future, particularly the theoretical variety. Proposed answers tend in the direction that scientists must increasingly focus on setting goals and asking the right questions, and that this may be the only thing left for them to do.

Cultural Factors and Training

I have always wondered about the mindset that an academic must have in order to indulge in misconduct. In my own institute I talked to a few students about it, and they indicated that cheating was widespread and tolerated in their schools. One student who we had to punish for copying in an exam even told me with disarming frankness: “Sir, when we joined this institute, you should have told us you were serious about ethical issues!” This suggests that our institutions are failing to convey the right message, even today.

Having delivered talks at academic institutes in India and conducted training sessions for students over several years, I found it surprisingly easy to impart an ethical understanding to them; indeed, they came to enjoy learning about the subject. Where the training was mandatory, they would come in with a slightly resentful air expecting to hear a bunch of homilies but were always pleasantly surprised. I would ask them: if your relative was to have surgery tomorrow and you found out today that the surgeon cheated on their medical school exam, how exactly would you feel? If your performance slipped by a grade because another student cheated his or her way to get ahead of you, how would that affect you? And if you wished to study abroad but your desired mentor was reluctant to admit an Indian due to a negative perception of our ethical standards, how would you react? 

These sessions brought home to students the reason why ethical standards work primarily in their own interest, and by the end of the session they were nodding enthusiastically. They realized that an unethical environment degrades trust all round, both within a peer group and from the perspective of external groups.

Explore the full InterDialogue with Sunil Mukhi.

But then I asked them a different question: if you became aware that a friend or colleague was cheating, would you call out that person, at least in private if not publicly? It became clear to me that there is a great reluctance to do so. In our society, “loyalty” is considered very important and—illogical as it may be—calling out an unethical friend or colleague is seen as disloyal. So students can expect to get away with misconduct as long as the authorities do not find out.

And here we must talk about the authorities. I have unfortunately encountered some academic leaders who have a thinly veiled hostility to the very existence of ethics guidelines and regulations. Far from being concerned about the kind of misconduct that is rampant in our country, even if it is brought to their attention, they seem to hunt for procedural obstacles to taking any action. The bogey of whether suspension from an academy could lead to lawsuits was brought up multiple times over many years. While the concern is perhaps justified to some extent, I felt it was being used as an argument to stall implementation. Ultimately, though, with changing times, avoidance is no longer an option, lawyers are being duly consulted, and there are welcome signs of improvement.

Responses to Ethical Misconduct

While ethical misconduct has regional and cultural flavors, the problem is of course global. The “Schön scandal” was committed by a German researcher working in the US, and led to his dismissal from Bell Laboratories as well as revocation of his PhD by the University of Konstanz. The physics department head was quoted as saying²⁹ “the case of Schön is the biggest misconduct scandal in physics in the last 50 years.” Also, some years ago, the president of Stanford University had to resign after the editor of a student magazine, Theo Baker, exposed research misconduct³⁰ in his work.

For me, what is notable in both cases is the seriousness and professionalism of the concerned institutions in investigating the allegations, publishing their findings transparently and punishing the perpetrator. In particular, the University of Konstanz invoked the Baden-Württemberg university law that “a doctoral degree can be retroactively revoked on grounds of dishonorable conduct” despite the misconduct not having been committed there. Similarly, Theo Baker has stated in a book that the Stanford University president’s resignation was (at least privately) demanded of him by the Board of Trustees of the university.

In India (and plausibly all of South Asia), seriousness and professionalism in the face of misconduct is often conspicuous by its absence. During the Rajput scandal detailed above, the governor of the state where the university was located as well as a student association on the campus issued press statements to the effect that no misconduct had been committed by him. They pointed to “vested interests” bent on defaming the state. Of course, they fell silent once the investigation was announced by the President of India and concluded with a guilty finding. Indians also rightly find it hard to imagine that a student magazine editor like Theo Baker, whose courageous investigations brought about the resignation of the president of Stanford, would be taken seriously or even be allowed to publish his findings if a similar case had arisen in India.

When leaders communicate indifference, they allow misconduct to grow and become rampant. One suspects that they are afraid of their own rules and procedures being applied to themselves, which they would see as an unacceptable challenge to their authority. 

On the flip side, some serious, committed academic leaders in India have worked to create an ethical environment in their institutions. When the leader states ethical ground rules explicitly and often, it can work wonders. It appears to me that younger generations of students and faculty show more concern about academic ethics than their seniors, and one can only hope that they will do better than preceding generations in addressing these issues once they become leaders.

Banner image: Sunil Mukhi delivers a talk about academic misconduct and AI ethics at the 17th General Conference of TWAS in October 2025. Credit: DH Rasolt.

1.  I must emphasize at the outset that I have looked into ethics from the standpoint of a teacher and administrator, but am not an expert or scholar in the field of ethics per se.
   ↩︎
2. Frontline. (2004). Vice-chancellor’s papers plagiarized, physicists allege ↩︎
3. The equivalent of a university president in the US. ↩︎
4. Alvarez-Gaumé, L., Freedman, D. Z., & Mukhi, S. (1981). The background field method and the ultraviolet structure of the supersymmetric nonlinear σ-model. Annals of Physics, 134(1), 85–109. ↩︎
5. Mukhi, S., & Vafa, C. (1993). Two-dimensional black hole as a topological coset model of c = 1 string theory. Nuclear Physics B, 407(3), 667. ↩︎
6. Mukhi, S., et al. (2002). Physics Plagiarism Alert [Archived webpage]. GeoCities. ↩︎
7. Times of India. (2002). Kumaon varsity V-C in plagiarism controversy. ↩︎
8. Frontline. (2004). Vice-chancellor resigns amid plagiarism findings. ↩︎
9. Indian Academy of Sciences. (2017). Scientific values: Ethical guidelines and procedures. ↩︎
10. National Medical Commission. (2002). Code of Medical Ethics Regulations, 2002. ↩︎
11. University Grants Commission (India). (2018). UGC (Promotion of Academic Integrity and Prevention of Plagiarism in Higher Educational Institutions) Regulations, 2018. ↩︎
12. Still not publicly available. ↩︎
13.  Tran, N. (2024). The ‘publish or perish’ mentality is fuelling research paper retractions at an alarming rate. The Conversation. ↩︎
14. Grimes, D. R., Bauch, C. T., & Ioannidis, J. P. A. (2018). Modelling science trustworthiness under publish or perish pressure. Royal Society Open Science, 5(1), Article 171511. ↩︎
15. Fanelli, D. (2016). Honest errors and self-retraction should not be stigmatized or penalized. Nature, 531(7595), 415. ↩︎
16. University of Oxford. Plagiarism. Study Skills and Training. Retrieved September 2026, from https://www.ox.ac.uk/students/academic/guidance/skills/plagiarism ↩︎
17. International Center for Academic Integrity. (n.d.). Graduate students’ knowledge about unintentional plagiarism. ↩︎
18. The World Academy of Sciences. Code of conduct on research integrity. ↩︎
19. The article does state that “publicly known incidents of sabotage in science are rare,” citing only one other similar case. ↩︎
20. American Association for the Advancement of Science. (2014). Sabotaged scientists. Science. ↩︎
21. To make a vague assertion here—based, as mentioned above, only on my reading of news reports without having access to the finer details—it appears that misconduct in the developed world may have a more individualistic and deliberate character as compared to the developing world, where it may be more systemic and involve sloppy ethics and herd behavior.
    ↩︎
22. Agrawal, A., Lakhotia, S. & Mukhi, S. (2024). India needs an autonomous research integrity office. The Indian Express. ↩︎
23. It is worth mentioning that the authors’ original title for this article was “Obsession with rankings is damaging India’s academic ecosystem.” ↩︎
24. Meho, L. I., & Akl, E. A. (2025). Using bibliometrics to detect questionable authorship and affiliation practices and their impact on global research metrics: A case study of 14 universities. Quantitative Science Studies, 6, 63–98. ↩︎
25. Wessel, M., et al. (2026). LLM use, cheating, and academic integrity in software engineering education. arXiv preprint 2603.17060v2. ↩︎
26. Guevara, A., et al. (2026). Single-minus gluon tree amplitudes are nonzero. arXiv preprint 2602.12176. ↩︎
27. OpenAI. (2026). Model disproves discrete geometry conjecture. ↩︎
28. Pedreschi, D., Barabási, A.-L., Eliassi-Rad, T., Vespignani, A., et al. (2025). Human-AI coevolution. Artificial Intelligence, 339, Article 104244. ↩︎
29. Physics Today. (2004). Schön’s physics PhD is revoked. ↩︎
30. PBS NewsHour. (2023). Student journalist discusses reporting that led to Stanford president’s resignation. ↩︎