The Scopus Burger Joint

Today, if a tree falls in a forest and no one indexes it in Scopus, did it really make a sound? Probably not, and certainly no accreditation committee would grant a five-year research bonus (quinquenio) for it. Welcome to the thrilling world of Academia. In this wonderful setting, we are witnessing a fascinating mutation of the classic “publish or perish” mantra into what we might call the birth of microwave scientific literature.

We are entering the era of microwave papers

Why spend years of your life collecting field data, dealing with real ethics committees, or funding expensive laboratories, when you can sit in front of a computer on a Friday afternoon and “cook up” a scientific article ready to submit on Monday?

The Miracle of the multiplication of the loaves and the papers

We are witnessing an unprecedented, exponential increase in the publication of systematic literature reviews (SLRs), meta-analyses, and, above all, bibliometric analyses. To give us an idea of the speed at which we are “producing science”, a recent study in educational research revealed that 87% of all bibliometric reviews indexed in Scopus in this field were published within a five-year period (between 2020 and 2024), and a staggering 97% since 2015 (link).

Overnight, methodologies based on collecting and repackaging what others have already written have become the dominant methodological approaches. We no longer research the world; we research what people say they research about the world. It is the epistemology of a mirror facing a mirror.

The New Scientific Geography (Or the revenge of the periphery)

And as if that were not enough, we have another component in this academic diet. Historically, the original knowledge club has been a private reserve for Anglo-American and Western European societies. But, surprise, surprise! In the realm of bibliometric reviews, the rules of the game have changed drastically.

Currently, 69% of bibliometric reviews published in education come from authors in emerging nations (Asia, Africa, and Latin America). In other words, non-Western researchers have become the official chroniclers of the decline of the empire of rigor, using the very same automated tools.

Why run when you can “automate”?

What explains this sudden obsession of emerging regions with bibliometrics and systematic reviews? It is not a collective intellectual epiphany, but simple academic Darwinian survival:

1. System Pressure: There is a global demand for researchers to publish at an industrial pace in high-impact journals. If you do not publish, you are out of the game.
2. The Technological “Fast Track” (The Quick Route): If your university lacks the budget for chemical reagents, telescopes, or anthropological fieldwork, free bibliometric software (like VOSviewer or Biblioshiny) is a godsend. It allows you to download a .csv file from Scopus, drag it into the software, and—voilà!—generate multicolored sociometric maps in three seconds that simulate “high science” without costing a single dime.
3. The Linguistic Scaffolding of AI: Writing in polished academic English no longer requires a Fulbright scholarship. Generative AI tools act as cognitive scaffolds, reducing writing anxiety and eliminating language barriers for non-native researchers. And this is not just for translation, as they can also be used to explain concepts, deduce ideas, propose analyses, etc.

The Dark Side of Template Science: pretty charts, empty brains

As expected, when you democratize the ability to manufacture articles effortlessly, methodological quality decides to jump out the window. Thus, it stands to reason that we are inundated by an “indiscriminate growth of low-quality studies” that barely surpass the minimum standards of scientific decency.

The symptoms of this academic pathology are obvious:

• Software-driven bias: Novice authors who confuse the automatic generation of word clouds and co-citation maps with real analysis. The result is a “serial presentation” of flashy charts accompanied by painfully superficial descriptions (along the lines of “Word X has a large node, but Word Y is in blue”) devoid of any deep analytical synthesis (neither deep, nor of any kind whatsoever).
• The clone factory: Multiple reviews are published on the same topic, in the same time frame, and using the exact same methods. Worst of all, these clones do not even cite each other! And sometimes they contradict one another.
• AI lobotomy (“permanent cognitive offloading”): Delegating writing and critical analysis to language models atrophies the ability to think independently (as if social networks, pro-government television, Bad Bunny, and reggaeton weren’t enough). Even worse, AI models are prone to suffering from “hallucinations”, inventing data, facts, and entire bibliographic references that later end up printed in “peer-reviewed” scientific journals.

Is there a cure for the bibliometric hangover?

If we want to rescue academia from becoming an automatic generator of indexed spam, we must apply urgent corrective measures:

• Penalize redundancy: Journals must immediately reject any review that does not demonstrate a critical knowledge gap or that simply repeats an analysis that was already done six months ago.
• Prioritize interpretation over software: Require that computer-generated maps be the starting point of a qualitative conceptual discussion, not the final result of the paper.
• Adopt rigorous evaluation criteria: Use formal rubrics that punish the lazy use of technology and reward true intellectual value.

To do our bit to avoid becoming “microwave academics”, we have prepared a condensed checklist for AI integration in academic analysis that we try to apply to our publications before “slipping them in” anywhere:

1. Ethical dimension (integrity filter): Have you manually verified every AI-generated citation and data point to ensure it is not a hallucination? Does the text retain your original voice and avoid the indirect plagiarism of synthetic structures?
2. Methodological dimension (rigor filter): Are you using the software to answer genuine research questions, or are you just “following the software” to fill pages with colored maps? Is there a qualitative synthesis that explains what the data networks mean beyond mere description?
3. Cognitive dimension (scaffolding vs. substitution): Did the AI act as a scaffold to relieve your mechanical load (grammar, basic structure), or did you use it to outsource critical thinking and the drafting of the core argument of your study?

Technology is an excellent crutch for working, but a terrible substitute for a thinking brain. In this era of quality crisis, true academic rebellion is not about publishing fifty articles a year, but about publishing a single one that is truly worth reading.