UNIVERSIDADE ESTADUAL PAULISTA

“JÚLIO DE MESQUITA FILHO”

Instituto de Ciência e Tecnologia

Campus de São José dos Campos

DOI: https://doi.org/10.4322/bds.2023.e3988

Braz Dent Sci 2023 July/Sept; 26 (3): e3988

EDITORIAL

Scientiﬁc writing with artiﬁcial intelligence: key considerations and alerts

Escrita cientíﬁca com inteligência artiﬁcial: principais considerações e alertas

How to cite: Fardim KAC, Gonçalves SEP, Tribst JPM. Scientic writing with articial intelligence: key considerations and alerts. Braz Dent Sci.

2023;26(3):e3988. https://doi.org/10.4322/bds.2023.e3988

ABSTRACT

The integration of articial intelligence (AI) text generators in scientic reports demands careful evaluation

of specic ethical considerations. While these AI technologies offer text generation support, addressing the

ethical implications is vital. This editorial highlights the need for a thoughtful and responsible approach,

emphasizing the establishment of guidelines and best practices by researchers and scientic communities.

Collaborative efforts between AI developers, researchers, and ethical committees can ensure the seamless

integration of AI technologies while upholding the integrity, quality, and ethical standards of scientic

reporting. This text comprehensively summarizes the key considerations to be followed when utilizing

articial intelligence text generators in scientic reports.

KEYWORDS

Articial intelligence; Dental society; Ethics; Forecasting; Scientic society.

RESUMO

A integração de geradores de texto de inteligência articial (IA) em relatórios cientícos exige uma

avaliação cuidadosa de considerações éticas especícas. Embora essas tecnologias de IA ofereçam

suporte à geração de texto, abordar as implicações éticas é fundamental. Este editorial destaca a

necessidade de uma abordagem ponderada e responsável, enfatizando o estabelecimento de diretrizes

e melhores práticas por parte de pesquisadores e comunidades cientícas. Esforços colaborativos

entre desenvolvedores de IA, pesquisadores e comitês éticos podem garantir a integração perfeita

das tecnologias de IA, ao mesmo tempo em que mantêm a integridade, qualidade e padrões éticos

da divulgação cientíca. Este texto oferece um resumo abrangente considerações-chave ao se utilizar

geradores de texto de inteligência articial em relatórios cientícos.

PALAVRAS-CHAVE

Inteligência articial; Sociedade odontológica; Ética; Previsão; Sociedade cientíca.

Braz Dent Sci 2023 July/Sept; 26 (3): e3988

Fardim KAC et al.

Scientiﬁc writing with artiﬁcial intelligence: key considerations and alerts

Fardim KAC et al. Scientiﬁc writing with artiﬁcial intelligence:

key considerations and alerts

CONTEXT AND PROSPECTS

Overcoming barriers and pushing beyond

a beautiful smile, modern dentistry subtly

consolidates itself, conquering millimeter by

millimeter new advances on the threshold of

knowledge. In this constant search for professional

assertiveness and patient well-being, performance

in diagnosis, treatment planning, improvement

of materials, harmonization, and execution of

techniques are the agships of studies in various

specialties [1-3]. Recent research explores

promising techniques, and thus, primordial

dentistry establishes its foundations in the “gold

standard” and underpins the steps of progress

in modern/contemporary dentistry, which now

goes hand in hand and is already awakening

with Articial Intelligence (AI) [4]. What once

seemed distant to us is now perceived as part

of our routine: Speech recognition and natural

language processing voice assistants provide

personalized and sometimes humorous responses

to our inquiries. They are capable of performing

daily tasks and controlling other smart devices.

Internet browsing behavior suggests products,

websites, and images based on previous searches,

feeding into this “intelligence”. Trafc control

systems optimize vehicle ow and safety, facial

recognition aids in identication, and big data

analysis targets patterns and trends, adding

significant value to areas such as marketing,

nance, arts (Figure 1), and scientic research.

AI silently permeates our daily lives.

Speaking of scientific research and the

promising capabilities of numerous AI tools,

we now turn to the production or renement

of scientic writing. While AI appears to assist

in the development of laboratory research, its

use for improving and perhaps even guiding

scientic language is a subject of interest [5].

Researchers and authors continue to play a

crucial role since no groundbreaking idea can

be developed without skillful and meticulous

hands and thoughtful minds. Creating a paper

using AI may seem akin to plagiarism, as it

relies on existing knowledge [6]. Therefore,

we acknowledge the merit of AI in language

corrections and improvements, which could

potentially minimize the “Tower of Babel” effect

between countries. However, it is essential to

emphasize that AI is powered by brilliant minds

who have worked naturally, posing questions

and searching for answers the “old way.” The

human mind is necessary for creative and

ethical thinking, while AI currently acts as a

helpful tool rather than a threat. It relies on

properly trained and qualied individuals to

function effectively.

ETHICAL CONSIDERATIONS IN THE

DENTAL FIELD

To provide valuable insights, this editorial

presents a summary of ethical considerations

within the dental field that can also apply to

health sciences. The following six topics are

described in detail below:

1. Accuracy and Reliability: One primary concern

is ensuring the accuracy and reliability of the

AI-generated text. Dentistry involves critical

decision-making processes that directly

impact patients’ oral health. Any inaccuracies

or errors in the AI-generated text could lead

to incorrect diagnoses, treatment plans, or

advice, potentially causing harm to patients.

2. Accountability and Responsibility: As AI

systems are developed and trained by

human creators, the issue of accountability

and responsibility comes into play. It is

essential to establish clear guidelines and

regulations to determine who is responsible

for the actions and decisions made based

on AI-generated text in dentistry. This

includes determining liability in case of any

negative consequences resulting from the

AI’s recommendations.

Figure 1 - AI-generated image using the input “open access science,

digital art”

Braz Dent Sci 2023 July/Sept; 26 (3): e3988

Fardim KAC et al.

Scientiﬁc writing with artiﬁcial intelligence: key considerations and alerts

Fardim KAC et al. Scientiﬁc writing with artiﬁcial intelligence:

key considerations and alerts

3. Informed Consent and Transparency: Dentists

have an ethical obligation to provide patients

with accurate and comprehensible information

about their oral health and treatment options.

With AI-generated text, it becomes crucial to

ensure that patients are fully informed about

the use of AI in their diagnosis or treatment.

Dentists should communicate the limitations,

potential biases, and uncertainties associated

with AI-generated text to obtain informed

consent from patients.

Data Privacy and Security: AI systems rely

heavily on vast amounts of data to train and

improve their performance. Dentistry involves

handling sensitive patient information, such as

medical records, X-rays, and personal details.

It is imperative to protect patient privacy and

ensure that AI-generated text is developed

and used in compliance with data protection

regulations. Safeguards should be in place to

prevent unauthorized access, use, or misuse of

patient data.

5. Equity and Bias: AI systems are susceptible to

biases present in the data they are trained on,

potentially resulting in unequal treatment or

disparities in dental care. To avoid perpetuating

existing biases, it is essential to address data

biases during the development and training of

AI systems. Efforts should be made to ensure

that the AI-generated text is fair, unbiased, and

equitable, providing equal and appropriate

care to all patients, irrespective of their

demographic characteristics.

Professional Autonomy and Human Judgment:

AI-generated text should be viewed as a tool to

augment human decision-making rather than

replacing it entirely. Dentists must retain their

professional autonomy and use AI-generated text

as an aid in their clinical practice, considering

it alongside their expertise and patient-specic

factors. Human judgment, empathy, and

intuition should continue to play a central role

in dentistry, ensuring the holistic care of patients.

Addressing these ethical considerations

requires a multidisciplinary approach involving

dental professionals, AI developers, regulatory

councils, and ethicists. It is essential to establish

clear guidelines, standards, and regulations to

govern the development, implementation, and

use of AI-generated text in dentistry, with the

primary aim of promoting patient well-being,

safety, and ethical practice.

WHY SHOULD WE CHECK IT TWICE?

Checking something twice is often done

to ensure accuracy, avoid errors, or maintain

quality. In the AI scenario the terms “garbage in,

garbage out” (GIGO) refer to the concept that

the quality of output or results from a system is

directly dependent on the quality of the input or

data provided to it [7]. It suggests that if you feed

a system with faulty, inaccurate, or low-quality

data, the output or results produced by the system

will also be awed, inaccurate, or of low quality.

“GIGO” highlights the importance of

ensuring high-quality input data and the careful

validation and preprocessing of data before using

it to make clinical decisions or use it in a scientic

manuscript. It serves as a reminder that even the

most advanced algorithms or technologies can

only work with the information provided to them

and are not capable of compensating for poor-

quality or awed data. Consequently, by ensuring

that high-quality data is fed into a system, we

increase the likelihood of obtaining meaningful

and trustworthy results [7,8].

Without checking the AI-provided information,

and trying to boost productivity without ethics,

authors can cite non-existing references and

made-up data from unreliable sources as noticed in

a retracted preprint article that contained several

fake references [9].

Understanding each AI tool and its applicability

can make its incorporation into daily life easier

and more useful. The language models generate

responses based on patterns and information

present in the training data it has been exposed

to, including information about the dental

eld [10]. In its current status, it does not have

real-time access to the internet and cannot

retrieve or verify the latest information [11]. The

model’s responses are based on statistical patterns

rather than a deep understanding of specic facts

or real-time events and therefore may produce

inaccurate information about people, places,

facts, or even references from the article.

GUIDELINES AND SUGGESTIONS FOR

SCIENTIFIC REPORTS

When using AI-generated text in scientic

reports, it is important to follow certain guidelines

to ensure responsible and ethical usage. Here are

some key guidelines to consider:

Braz Dent Sci 2023 July/Sept; 26 (3): e3988

Fardim KAC et al.

Scientiﬁc writing with artiﬁcial intelligence: key considerations and alerts

Fardim KAC et al. Scientiﬁc writing with artiﬁcial intelligence:

key considerations and alerts

Transparency and disclosure: Indicate when

AI-generated text has been used in the

report. Provide information about the specic

AI models, algorithms, or tools employed, as

well as the training data sources. Transpar-

ency helps readers understand the potential

limitations, biases, and uncertainties associ-

ated with AI-generated text.

Verication and validation: Validate the accu-

racy, reliability, and relevance of AI-generated

text before incorporating it into scientific

reports. Cross-check the generated content

against trusted sources, conduct additional

research or experiments to conrm ndings,

and critically evaluate the outputs to ensure

they align with scientic rigor. Open-access sci-

ence can play an important role in this regard

[11].

Contextualization and interpretation: Provide

proper context and interpretation for AI-

generated text. Clearly distinguish between

human-authored content and AI-generated

content. Explain the purpose and limitations

of AI-generated text, and offer critical analysis

and interpretation to complement the gener-

ated results.

Collaboration between AI and human research-

ers: Recognize that AI is a tool that can assist

in scientic research but should not replace

human expertise. Foster a multidisciplinary

approach where researchers with domain

knowledge work alongside AI systems, com-

bining their respective strengths for more

robust scientific reports. Using AI to help

write a scientic article is not equivalent to

co-authorship. AI lacks the ability to contribute

original ideas, understand context, make ethi-

cal decisions, or participate in rigorous critical

review. It can only be considered a helpful

tool, but not a co-author. Mentioning AI as

a tool is acceptable, but it cannot be granted

co-author status [12].

5. Ethical considerations: Be mindful of the

ethical considerations surrounding AI-gener-

ated text, as discussed earlier. Address issues

such as bias, misinformation, intellectual

property, privacy, and accountability. Take

steps to mitigate biases, ensure accuracy,

protect privacy, and properly attribute

sources.

Peer review and validation: Submit AI-gener-

ated text within scientic reports to rigorous

peer review. Independent experts can assess

the scientic validity, ethical implications,

and appropriateness of using AI-generated

text. Peer review helps maintain the quality

and integrity of scientic reports and pro-

vides additional scrutiny for AI-generated

content.

7. Compliance with publishing guidelines: Fol-

low the publishing guidelines of the target

scientic journals or venues when including

AI-generated text. Ensure compliance with

citation standards, disclosure requirements,

and any specic guidelines related to the use

of AI or machine-generated content.

Ongoing monitoring and improvement:

Continuously monitor and evaluate the per-

formance and impact of AI-generated text.

Identify and address potential biases, inaccu-

racies, or unintended consequences that may

arise. Regularly update and rene AI models,

algorithms, and training data to improve the

quality and reliability of the generated text.

By adhering to these guidelines, researchers

can incorporate AI-generated text responsibly

into scientic reports while upholding scientic

integrity, transparency, and ethical principles.

Acknowledgements

The authors would like to acknowledge the

utilization of ChatGPT, an AI language model

developed by OpenAI, that provided initial

drafts and suggestions, which were subsequently

revised and improved by the authors. The English

language correctness and overall coherence of

the text were veried through the collaboration

between the authors.

The figure presented in this editorial was

generated using DALL-E, an AI model developed

by OpenAI that specializes in creating images

from textual descriptions. DALL-E utilizes a

combination of deep learning and generative

modeling techniques to produce unique and novel

visual representations based on the input provided.

Author’s Contributions

KACF: Conceptualization, Writing, Original

Draft Preparation, Review & Editing. SEPG:

Original Draft Preparation, Writing, Review &

Braz Dent Sci 2023 July/Sept; 26 (3): e3988

Fardim KAC et al.

Scientiﬁc writing with artiﬁcial intelligence: key considerations and alerts

Fardim KAC et al. Scientiﬁc writing with artiﬁcial intelligence:

key considerations and alerts

Editing. JPMT: Conceptualization, Original Draft

Preparation, Supervision, Review & Editing.

References

1. Arcila LV, Gomes LC, Travassos A, Fernandes VV, Nascimento RD,

Ursi W, et al. Multidisciplinary action in a patient with cleft

lip and palate: esthetic-functional dentofacial treatment for

more than two decades. Braz Dent Sci. 2023;26(2):e3801.

http://dx.doi.org/10.4322/bds.2023.e3801.

2. Silva TM, Faria Petrucelli N, Mendonça RP, Silva JP Jr, Campos

TM, Paiva Gonçalves SE. Impact of photoinitiator quality on

chemical-mechanical properties of dental adhesives under

different light intensities. Braz Dent Sci. 2023;26(1):e3704.

http://dx.doi.org/10.4322/bds.2022.e3704.

3. Andrade GS, Kalman L, Giudice RL, Adolfi D, Feilzer AJ, Tribst JP.

Biomechanics of implant-supported restorations. Braz Dent Sci.

2023;26(1):e3637. http://dx.doi.org/10.4322/bds.2023.e3637.

4. Paridhi A, Pradnya N. Artificial Intelligence in dentistry: past,

present, and future. Cureus. 2022;14(7):e27405. PMid:36046326.

5. Ding H, Wu J, Zhao W, Matinlinna JP, Burrow MF, Tsoi JK.

Artificial intelligence in dentistry: a review. Frontiers in Dental

Medicine. 2023;4:1085251. http://dx.doi.org/10.3389/

fdmed.2023.1085251.

6. Elali FR, Rachid LN. AI-generated research paper

fabrication and plagiarism in the scientific community.

Patterns. 2023;4(3):100706. http://dx.doi.org/10.1016/j.

patter.2023.100706. PMid:36960451.

7. Bignami EG, Vittori A, Lanza R, Compagnone C, Cascella M,

Bellini V. The clinical researcher journey in the Artificial Intelligence

Era: the PAC-MAN’s challenge. Healthcare. 2023;11(7):975.

http://dx.doi.org/10.3390/healthcare11070975. PMid:37046900.

8. Rodrigues JA, dos Santos HD. Is artificial intelligence a future

trend in health care? Braz Dent Sci. 2021;24(3)

9. Joelving F. Publisher blacklists authors after preprint cites

made-up studies [Internet]. 2023 [cited 2023 Aug 7].

Available from: https://retractionwatch.com/2023/07/07/

publisher-blacklists-authors-after-preprint-cites-made-up-studies/

Sergio Eduardo de Paiva Gonçalves 

Editor in chief

São Paulo State University – UNESP,

Restorative Dentistry Department, São José

dos Campos, São Paulo, Brazil.

João Paulo Mendes Tribst 

Section Editor

Universiteit van Amsterdam and Vrije

Universiteit Amsterdam, Academic Centre

for Dentistry Amsterdam, Department

of Oral Regenerative Medicine, LA

Amsterdam, The Netherlands.

Karolina Aparecida Castilho Fardim 

Editor Assistant

São Paulo State University, Department

of Diagnosis and Surgery, São José dos

Campos, SP, Brazil.

10. Eggmann F, Blatz MB. ChatGPT: chances and challenges for

dentistry. Compend Contin Educ Dent. 2023;44(4):220-4.

PMid:37075729.

11. Tribst JPM, Dal Piva AMDO, Kleverlaan CJ, Feilzer A. Is there

still a need for dental textbooks? When doubts can be

clarified by artificial intelligence and open access science.

Front Dent Med. 2023;4:1236861. http://dx.doi.org/10.3389/

fdmed.2023.1236861.

12. Thorp HH. ChatGPT is fun, but not an author. Science.

2023;379(6630):313. http://dx.doi.org/10.1126/science.

adg7879. PMid:36701446.