At C2C Journal, Ottawa-based writer and lawyer Lynne Cohen offers a long and very informative article on the problem of academic fraud.

Her comments on the role that chatbots have begun to play in academic fraud are among many that are worth noting:

Adding to the longstanding and perverse incentives of the academic world is another, new threat to research excellence: artificial intelligence. For academics under pressure to produce research and tempted by the prospect of cutting corners or even cheating outright to win plaudits, AI offers an apparent godsend. The ability to produce complete (but phony) articles at the figurative push of a button threatens to multiply academic fraud by orders of magnitude…

As an Agence France-Presse (AFP) report entitled “Flood of ‘junk’: How AI is changing scientific publishing” explained last year, AI has already generated a long list of absurdities that have appeared in published academic papers. This includes a graphic of a rat with ludicrously large genitals, a picture of human legs with too many bones, and a paper that begins suspiciously with: “Certainly, here is a possible introduction for your topic.” While such examples may seem easy to spot, as AI improves, such errors are likely to disappear as the fakes become ever-more convincing. The AFP report argues that AI has “turbocharged” the problem of academic fraud. This may be the case, but it is important to remember that while technology can facilitate bad behaviour, it can’t create the initial impulse to cheat. “I think a focus on ChatGPT misses the point,” advises [Retraction Watch’s Ivan] Oransky “ChatGPT has really just industrialized what is already happening by making it easier to…corrupt the scientific record.”

He doesn’t dispute that AI is an issue, but argues it’s “a symptom of the bigger problem”. The fact that no field appears to be immune from scandalous frauds – plagiarism, bogus findings and fakery afflict not only the natural or “hard” sciences but social sciences, humanities and all other academic areas – plus the fact that it occurs worldwide, suggests that this is fundamentally a problem of human behaviour rather than technology. The only way to tackle the problem, Oransky believes, is to figure out how to “get rid of bad incentives in science.” And the first step is to make it unprofitable for individuals or institutions to produce bad science.

“Research to Ruin: The Worsening Spectre of Academic Fraud,” January 21, 2025

That’s the hard part .