Serendipity Revisited: What 533 Nobel Prizes Reveal About Breakthrough Science

A recent paper published in Scientometrics explores whether scientific breakthroughs—including those often labeled as “serendipitous”—can be explained in causal and statistical terms. The study tests a fundamental hypothesis: whether there is an underlying logical structure to how discoveries emerge, rather than breakthroughs arising purely by chance.

The analysis covers science’s most influential and canonized discoveries, encompassing all 533 Nobel Prize–winning discoveries from the prize’s inception in 1901 through 2022. Given that not all major discoveries receive a Nobel Prize, the study also examines landmark discoveries documented in leading science textbooks, including “top 100” lists of the greatest scientists and their contributions across disciplines and historical periods. The central conclusion is that scientific discovery is far less random than commonly assumed. What are often described as “serendipitous” breakthroughs are, in most cases, enabled by the development and application of new tools and methods, rather than by chance alone. These tools create the conditions in which unexpected findings become possible, repeatable, and increasingly likely.  https://link.springer.com/article/10.1007/s11192-025-05503-y

But if future breakthroughs depend more on tools than on luck, does that mean the future of science is being shaped less by curiosity and more by whoever controls the money, the machines, and the infrastructure, often quietly and without public accountability, deciding what we get to discover and what stays out of sight? And as new tools set the limits of what can be known, will the things we never discover be written off as bad luck or seen for what they are, the result of choices about what gets built, shared, or kept out of reach, with lasting uneven consequences, for who wins, who loses, and whose realities never get noticed?

PS - I find it unfortunate that A. Krauss, the author of the aforementioned study did not cite a closely related study published in Nature Communications, discussed in my previous post, “The Hidden Equations Behind Scientific Progress: The Art of Engineering Serendipity.” That work reinforces the central message that even the most surprising scientific breakthroughs are not acts of pure chance, but emerge from deeper regularities that can be understood, anticipated, and shaped through causal and statistical insight. https://19-pacheco-torgal-19.blogspot.com/2025/02/the-hidden-equations-behind-scientific.html

O jornal Público volta hoje a dar destaque a um ranking universitário da pura treta

Infelizmente, o jornal Público volta hoje a dar destaque a um ranking universitário da pura treta. O mais irónico é que isso seja feito pelo mesmo jornal que, há poucos meses atrás, e de forma negligente, nada noticiou sobre as instituições de ensino superior portuguesas que conseguiram o milagre de ultrapassar as melhores universidades da Alemanha em cinco áreas científicas no prestigiado ranking Shanghai por áreas de 2025. Um resultado extraordinário, que deveria ter suscitado orgulho coletivo e ajudado a promover uma reflexão urgente sobre o enorme esforço desenvolvido, e a estratégia bem-sucedida dessas áreas científicas e a possibilidade de replicar esse modelo noutras áreas científicas, com vista a elevar a competitividade da ciência portuguesa. Em vez disso, aquele jornal opta agora por dar visibilidade a um ranking de qualidade profundamente duvidosa, com elevado potencial para confundir e desinformar os leitores, em particular aqueles muitos que não possuem conhecimentos mínimos sobre rankings universitários, pelo que mais não me resta assim do que reproduzir abaixo o mesmo comentário contundente que já havia feito num post anterior:

"Mais uma vez o jornal Público não se coibe de promover hoje um ranking da treta, o  QS World University Rankings.  Ou talvez no jornal Público não saibam que o ranking Shanghai, o único ranking a nível mundial que contabiliza os prestigiados prémios Nobel, é o único ranking mencionado num documento da Comissão Europeia sobre excelência científica. 

Ou talvez no jornal Público não saibam aquilo que conhecidos académicos, como o conhecido professor catedrático David Blanchflower, ou o professor catedrático da universidade de Oxford Simon Marginson, escreveram, em termos nada elogiosos sobre o ranking QS, reproduzidos na parte final deste post.

 
E quem não se lembra daquele outro ranking da treta, também noticiado pelo mesmo jornal Público, que dizia que a Católica era a melhor universidade Portuguesa ranking esse que o então Presidente da A3ES, o catedrático Alberto Amaral, reduziu a pó em dois artigos no Expresso, lembrando que os critérios relativos a citações desse famigerado ranking até colocavam universidades do Irão, do Egito e da Jordânia à frente da própria Universidade de Cambridge. 

 
Porém ainda antes mesmo dos referidos dois artigos do catedrático Alberto Amaral terem sido publicados no Expresso eu enviei na altura um email a alguns milhares de Colegas com o sugestivo título "UCatólica é a melhor universidade Portuguesa____Fake news, incompetência ou ambas?". https://www.docdroid.net/S7WaBal/fake-news-incompetencia-ou-ambas-docx
 
O mais irónico é que o mesmo jornal Público que costuma dar destaque a rankings da treta disse zero sobre o ranking Shanghai por áreas científicas divulgado recentemente https://www.jn.pt/local/noticias/braga/braga/uminho-entre-as-melhores-universidades-do-mundo-em-18-areas-cientificas-12446554.html  o que suscita legitimas dúvidas sobre quais são realmente os critérios jornalisticos do jornal Público. Se os do rigor jornalistico como apregoam aos quatro ventos ou os da desinformação encapotada. Uma coisa é certa como assinante do jornal Público considero uma vergonha pagar para ser "informado" sobre rankings que valem pouco mais do que lixo. 
 
PS - É importante recordar que o famigerado QS World University Rankings é produzido pela firma Quacquarelli Symonds, que foi fundada por um Italiano espertalhaço de nome Nunzio Quacquarelli, quando andou a fazer o seu MBA, firma essa que ganha milhões a vender (aos incautos) estrelas e outros serviços de aconselhamento, sobre como subir nos rankings. Vide email de 2018, https://www.docdroid.net/uniDTYH/vice-reitor-docx onde comentei o facto da Universidade de Coimbra ter sido um dos pagantes desse caro serviço, apesar de muito ironicamente isso não ter impedido logo a seguir que essa universidade caisse no ranking Shanghai. Email esse que na altura até foi divulgado pelo Carlos Fiolhais no seu blog.

Selecção de comentários de várias académicos sobre o ranking (da pura treta) QS: 

-David Blanchflower in an article for the New Statesman entitled "The QS Rankings are a load of old baloney"
"This ranking is complete rubbish and nobody should place any credence in it.The results are based on an entirely flawed methodology that underweights the quality of research and overweights fluff"
 
-Simon Marginson, professor of higher education at University of Oxford:
"I will not discuss the QS ranking because the methodology is not sufficiently robust"
 
-Fred L. Bookstein, Horst Seidler, Martin Fieder and Georg Winckler in the journal Scientometrics:
"There are far too many anomalies in the change scores of the various indices"
 
-Isidro F. Aguillo, Judit Bar-Ilan, Mark Levene, José Luis Ortega in the journal Scientometrics:
"The QS is based on a not large and not representative enough survey that means the results are biased towards certain countries"
 
-H. Jons and M. Hoyler in the Geoforum; Journal of Physical, Human, and Regional Geosciences
"The QS ranking was also criticized for the low response rates of the review surveys and for a general lack of methodological transparency"
 
-V. Safon in the journal Scientometrics:
"the majority of the received questionnaires come from English-speaking countries, clearly favoring their universities"

-Mu-Hsuan Huang in the journal Research Evaluation:
"the statistic data adopted by QS Rankings should be further questioned."
 
Andrejs Rauhvargers in Global University Rankings and their Impact- Report II:
"QS admits that a university may occasionally be nominated as excellent and ranked in a subject in which it “neither operates programmes nor research”

The anatomy of a highly cited ChatGPT paper and how it gained more than 3,000 Scopus citations in less than three years

A search in Scopus for publications with “ChatGPT” in the title, abstract, or key-words returns 28,562 documents published since 2022. Within this rapidly expanding body of literature, the second most cited publication is not an empirical study but an opinion paper published in 2023 by more than 70 authors from over 20 countries, titled “So what if ChatGPT wrote it? Multidisciplinary perspectives on opportunities, challenges and implications of generative conversational AI for research, practice and policy.” https://www.sciencedirect.com/science/article/pii/S0268401223000233 

That this paper has achieved such prominence within a corpus of more than 28,000 studies is not merely surprising; it is revealing of the field’s early intellectual dynamics. Although informed by a review of 364 publications and a broad international authorship, it introduces no new empirical evidence, testable propositions, or methodological advances, offering instead a wide-ranging interpretive synthesis produced at a moment when ChatGPT’s capabilities, uses, and institutional consequences were still rapidly evolving. 

Despite these limitations, the paper has been elevated to cross-disciplinary authority, with citations spanning nearly all major fields—from social sciences and computer science to business, engineering, medicine, mathematics, and the humanities—allowing a provisional narrative to circulate across communities with markedly different standards of evidence and, in many cases, to substitute for empirical grounding that was not yet available.

The paper enumerates an extensive set of risks and opportunities—spanning education, labour, cybersecurity, bias, and governance—without prioritising their importance, assessing relative likelihood or severity, or translating concerns into concrete policy frameworks. Readers are thus left with a catalogue of what might matter rather than guidance on what matters most. Moreover, many of the issues framed as disruptive simply repackage long-standing debates associated with earlier digital technologies—such as automation-driven job displacement, plagiarism, and misinformation—yet the paper fails to clearly separate what is genuinely new about large language models from familiar cycles of technological alarmism.

PS - In late January 2026, OpenAI’s CEO publicly acknowledged that a recent iteration of ChatGPT had sacrificed writing quality in favor of technical performance. This admission highlights a broader tension: much of the literature that quickly became authoritative was produced while the technology itself remained unstable, with regressions recognized by its developers. The speed with which normative interpretations solidified stands in contrast to the absence of empirically grounded criteria for evaluating capabilities that were still in flux.