This study investigates the adoption, perceived efficacy, and ethical considerations surrounding the use of AI-

powered tools in academic writing and peer review among academics in Nigerian Universities. Employing a

descriptive survey design, data were collected from 425 respondents across disciplines using structured

questionnaires. The findings reveal a moderate adoption rate of 62.12%, with generative AI tools being less

frequently used compared to grammar and writing assistance tools. While 88% of respondents perceived AI

tools as highly useful—particularly in enhancing writing quality—only 25% reported having the necessary

support to ensure ethical, responsible, and effective integration of AI tools in academic settings.

integrity

The emergence of artificial intelligence (AI) has fundamentally reshaped various aspects of scholarly

communication, particularly in academic publishing and peer review. AI-powered tools especially advanced

large language models like ChatGPT and similar systems—are now widely deployed to support tasks ranging

from language editing, plagiarism detection, citation formatting, and even preliminary manuscript evaluations

(Kokol et al., 2023; Lee et al., 2022; Floridi & Cowls, 2019). These tools have significantly enhanced the

efficiency and accuracy of research workflows, transforming how scholars search for literature, draft

landscape (Boden et al., 2021; Ienca & Vayena, 2020). In the Nigerian context, the use of AI in academic

institutions is still emerging. Research shows that most universities in sub-Saharan Africa, including Nigeria,

lack formal frameworks for the ethical and strategic use of AI in education and research (Okoye &

Nwachukwu, 2022; Obi & Alade, 2021). This policy vacuum creates inconsistencies in how AI tools are

adopted, potentially exposing institutions to academic integrity risks and widening the digital divide among

faculty and students (Adeleke, 2022; Bello & Asogwa, 2022). Ahmadu Bello University (ABU), Zaria one of

Nigeria’s foremost research institutions—stands at a critical juncture. With the increasing accessibility of

generative AI tools, Nigeria has an opportunity to lead by example in shaping context-specific AI governance

that reflects its infrastructural capacities, academic culture, and institutional values. Such proactive

engagement is essential to ensure that the adoption of AI enhances, rather than undermines, the quality,

technical support, inconsistent training on emerging technologies, chronic funding deficits, and unreliable

internet connectivity. Specifically in Nigeria, infrastructural challenges hinder students’ ability to effectively

utilize digital resources. Varying levels of digital literacy among faculty and students further exacerbate the

problem. For instance, Nigeria’s ICT policy mandates introductory computing courses and micro-certification

programs for undergraduates to promote digital literacy, broader studies indicate that many Nigerian university

students still lack the technical skills necessary to use digital tools and platforms effectively (Adeleke, 2022;

Bello & Asogwa, 2022). Moreover, university teachers in Nigeria often report inadequate training and

institutional support, particularly in relation to AI integration. This gap between basic digital competence and

AI-specific proficiency poses a major obstacle to meaningful and responsible adoption. The widespread

absence of formal institutional policies governing AI use in Nigerian universities creates an environment of

uncertainty and inconsistency. In the absence of clear, context-specific guidelines, academics in Nigeria are

left to navigate complex ethical and practical challenges alone—ranging from unaddressed plagiarism and

associated risks. Ultimately, this could compromise both academic integrity and the quality of scholarly work

produced within the institution.

review among academics in Nigerian universities?

2. How do perceived usefulness and ease of use influence the acceptance and continued use of AI-powered

tools for academic writing and peer review among academics in Nigerian universities?

3. How do academics in Nigeria perceive the efficacy of AI-powered tools in enhancing the quality, efficiency,

and integrity of academic writing and peer review processes?

4. What ethical concerns do academics in Nigerian universities, associate with the use of AI-powered tools in

academic writing and research?

3. To assess academics’ perceptions of the efficacy of AI-powered tools in improving the quality, efficiency,

4. To identify and analyze ethical concerns associated with the use of AI-powered tools in academic writing

and research as perceived by academics in Nigerian universities.

Artificial Intelligence (AI) tools have become increasingly integral to academic writing and peer review,

offering specialized support across various stages of scholarly work. These tools range in their functionalities,

aiding literature discovery, research design, manuscript development, data analysis, plagiarism detection, and

peer evaluation. For literature survey and research discovery, tools like Semantic Scholar, Iris.ai, Connected

Papers, Litmaps, and Zotero are commonly used. These platforms enhance the efficiency of data collection,

analysis, and reference management by enabling researchers to identify relevant studies, extract key findings,

and visualize the interconnection of scholarly works, as noted by Ige and Shorunke (2023). When it comes to

Grammarly, ProWritingAid, Hemingway Editor, QuillBot, and Trinka AI helps writers enhance grammar,

content creation are supported by applications like Tableau, Infogram, and Canva, which are widely utilized to

& Abubakar, 2022). To uphold academic integrity, plagiarism detection tools such as Turnitin AI, GPTZero,

help safeguard against both intentional and inadvertent plagiarism, as discussed by Bello and Hassan (2022).

In peer review processes, platforms such as Perplexity AI, Scite, Scholarcy, and Peerceptiv have proven useful

credibility (Adebayo & Johnson, 2023). It is important to distinguish between assistive AI tools, like grammar

checkers, and generative AI models, such as ChatGPT. The latter are designed to predict subsequent words in a

sequence rather than to ensure factual correctness. This predictive nature can lead to inaccuracies, omissions,

or fabricated content. Consequently, not all AI tools can be considered equally reliable or ethically suitable for

every academic task, a point underscored by Eze (2022).

The adoption of AI tools is best understood through established technology adoption models. The Technology

Acceptance Model (TAM) posits that an individual's intention to use a given technology is shaped by

perceived usefulness—defined as the belief that the technology enhances performance—and perceived ease of

and behavioral intentions, making TAM a valuable framework for examining the acceptance of AI tools

The level of AI tool awareness and adoption in Nigerian academia presents a mixed yet gradually evolving

landscape. Surveys show that while over 75% of Nigerian university students use AI tools primarily for

grammar correction and sentence restructuring, there remains a significant portion of the academic community,

particularly among library personnel and teaching staff, who demonstrate limited awareness and preparedness

to integrate AI tools into their work (Okoye & Lawal, 2022; Ibrahim, 2023). Among university librarians,

however, more than half report awareness of AI tools and express a strong willingness to adopt them.

Nonetheless, institutional barriers, such as inadequate digital infrastructure and the absence of enabling

policies, continue to impede broad-based integration (Bello & Hassan, 2022).

still lack the technical skills required to use AI tools effectively. Faculty members, too, often cite a lack of

adequate training in the use of AI for academic purposes (Eze, 2022). Furthermore, user perceptions of

usefulness and ease of use play a decisive role in adoption decisions. Male users frequently highlight the

performance advantages of AI tools, while female users tend to appreciate their user-friendliness. Across the

board, improvements in grammar, clarity, and ease of use are cited as major incentives for adoption (Okoye &

Lawal, 2022). Social influence and habitual use also drive adoption; peer recommendations and consistent

usage patterns make it more likely that these tools become embedded in daily academic routines (Adebayo &

Johnson, 2023). However, the absence of formal institutional policies remains a significant impediment.

Without structured ethical and operational guidelines, AI tool usage is often informal, inconsistent, and at

generate faulty citations and are generally ill-equipped to evaluate originality or engage in critical thinking—

capabilities that are fundamental to scholarly work (Eze, 2022). Evaluating the efficacy of these tools requires

a comprehensive approach that includes both quantitative and qualitative metrics. Quantitative measures such

as time saved, precision, and accuracy should be considered alongside qualitative indicators like user

satisfaction and perceived educational benefits. A blended evaluation method ensures a more holistic

understanding of the tools' impact (Bello & Hassan, 2022).

The rise of AI in academic settings introduces complex ethical challenges, particularly in areas like plagiarism,

authorship, and originality. AI-generated content can sometimes bypass traditional plagiarism detection

systems while lacking genuine originality. Moreover, AI tools cannot be held accountable for their output and

therefore cannot meet authorship criteria. An over-reliance on AI could potentially erode intellectual rigor and

AI assistance (Eze, 2022).

The study adopted a quantitative research approach. This approach allows for a comprehensive understanding

gather data on AI tool awareness, usage patterns, and perceived usefulness/ease of use from a broad sample of

usefulness, perceived ease of use, performance expectancy, effort expectancy, social influence, facilitating