ITA 2025 – Quais asserções sobre detecção de textos de IA são verdadeiras
As scientists explore AI-written text, journals hammer out policies Many ask authors to disclose use of ChatGPT and other generative artificial intelligence
“It’s all we’ve been talking about since November,” says Patrick Franzen, publishing director for SPIE, the international society for optics and photonics. He’s referring to ChatGPT, the artificial intelligence (AI) - powered chatbot unveiled that month. In response to a prompt, ChatGPT can spin out fluent and seemingly well-informed reports, essays — and scientific manuscripts. Worried about the ethics and accuracy of such content, Franzen and managers at other journals are scrambling to protect the scholarly literature from a potential flood of manuscripts written in whole or part by computer programs.
Some publishers have not yet formulated policies. Most of those that have avoid an outright ban on AI-generated text, but ask authors to disclose their use of the automated tools, as SPIE is likely to do. For now, editors and peer reviewers have few alternatives, as they lack enforcement tools. No software so far can consistently detect the synthetic text the majority of the time.[. . . ]
In some cases, the resulting text is indistinguishable from what people would write.
For example, researchers who read medical journal abstracts generated by ChatGPT failed to identify one-third of them as written by machine, according to a December 2022 preprint. AI developers are expected to create even more powerful versions, including ones trained specifi-cally on scientific literature - a prospect that has sent a shock wave through the scholarly publishing industry.
So far, scientists report playing around with ChatGPT to explore its capabilities, and a few have listed ChatGPT as a co-author on manuscripts. Publishing experts worry such limited use could morph into a spike of manuscripts containing substantial chunks of AI-written text.
Fonte: BRAINARD, Jeffrey. As scientists explore AI-written text, journals hammer out policies. Science, v. 379, n. 6634, p. 740-741, 22 fev.2023. Disponível em https://www.science.org/content/article/scientists-explore-ai- written-text-journals-hammer-policies.
Fonte: Science, 2023-02-22.
Leia as asserções destacadas e, em seguida, assinale a alternativa CORRETA. I. Pesquisadores acostumados a ler artigos de periódicos médicos não conseguem compreender um terço das publicações. II. Por ora, editores e revisores de publicações científicas têm dificuldades para iden- tificar, de forma consistente, textos gerados por IA dada a escassez de softwares especializados. III. Responsáveis por publicações científicas estão relutantes em coibir o uso de geradores de texto por computador, temendo uma potencial enxurrada de críticas por parte da comunidade acadêmica. IV. A publicação de prospectos gerados por IA, contendo novas versões de alguns artigos científicos específicos, causou uma onda de choque na indústria editorial acadêmica.
- A) Todas as asserções são verdadeiras.
- B) Somente as asserções II e III são verdadeiras.
- C) Apenas a asserção II é verdadeira.
- D) Somente as asserções III e IV são verdadeiras.
- E) Nenhuma das asserções é verdadeira.