Artificial intelligence in research: experts urge integration with human skills
Specialists have urged the development of artificial intelligence in research and its deliberate linkage to human skills and critical thinking, according to a lecture series organized by the Zoom team at the Omani Sociologists Association. The series examined how AI can be guided to serve knowledge production, support scholarly inquiry and analyze social behavior across communities and institutions.
Speakers said the technology is already embedded in educational and research settings, and institutions must adapt policies and practices to ensure AI tools strengthen, rather than replace, scientific judgment and methodological rigor.
How specialists framed the role of AI tools in scholarly work
During the talks, Dr. Fatima Al-Douhani, a certified trainer in institutional maturity for AI and future tools, described artificial intelligence in research as part of a broader digital ecosystem. She emphasized that these systems learn from data and mimic certain human capabilities, and that researchers need to shift from solitary workflows to collaborative, technology-informed partnerships.
Dr. Al-Douhani noted that AI tools should be considered supportive instruments that help organize tasks, reduce repetitive workload and accelerate access to literature and analysis. However, she stressed that machine assistance cannot substitute the researcher’s scientific reasoning and that foundations in research methodology remain essential.
Ethical risks and threats to critical thinking
Speakers highlighted ethical challenges associated with widespread AI adoption, warning against over-reliance on generated content. Experts cautioned that automated outputs can appear authoritative while containing inaccuracies or referencing non-existent sources, increasing the risk of academic plagiarism or misleading citations.
Furthermore, Dr. Al-Douhani and other panelists argued that excessive dependence on automation may weaken critical thinking and analytic skills if researchers treat tools as final arbiters of truth. Therefore, institutions and supervisors should require human verification and insist on preserving each researcher’s intellectual voice.
Building integrated research systems with AI
Khalil Al-Abdaly, founder of an academy for AI technologies, recommended moving beyond ad hoc tool use toward designing repeatable research systems. He said that a robust system combines a precisely framed research question, structured searches, verification steps, source management and clear prompts for AI components.
From prompts to workflow design
Al-Abdaly explained that productivity gains depend less on the number of tools and more on workflow architecture. Effective prompts should include role definition, context, information sources, scope, step sequence and output format. Meanwhile, more advanced agents can access files and web services to perform multi-step tasks, expanding the practical utility of AI in complex projects.
According to the speakers, the true value of artificial intelligence in research lies in organizing source collection, categorizing evidence, mapping connections across data and producing draft visualizations that accelerate human analysis.
Training researchers to preserve judgment and boost productivity
Dr. Ahmed Maher Shehata, an associate professor in information studies, argued that AI has become a partner in knowledge production by automating routine chores and enabling faster discovery. He observed that generative AI now produces text, images and data summaries that researchers can adapt for reports and presentations.
However, Dr. Shehata reiterated that the researcher remains responsible for accuracy, methodological consistency and ethical standards. He recommended training programs that combine digital literacy with classical research skills so that scholars learn to design effective prompts, verify AI outputs and avoid granting excessive access to sensitive data or systems.
Practical challenges: language, volume and digital equity
Panelists identified several practical barriers that AI must help overcome. One persistent issue is the sheer volume of new studies published daily, which complicates retrieval of the most relevant literature. Additionally, the dominance of English in peer-reviewed publications limits access for non-English-speaking researchers, making translation and summarization features desirable in AI tools.
Speakers also warned of a widening gap between communities that develop new technologies and those that primarily consume them, with implications for digital justice and equal access to research capabilities. They stressed institutional responses to ensure equitable deployment of AI and investment in capacity building across regions.
Policy implications and institutional responses
Experts called for institutional frameworks to govern the use of AI tools, including guidelines on authorship, disclosure of AI-assisted drafting, data privacy and algorithmic bias. They recommended multidisciplinary research agendas that combine computer science, social science and ethics to study AI’s societal effects and to develop transparent evaluation methods.
Educational programs should update curricula to emphasize analytical reasoning, data literacy and responsible tool use, according to the presenters. Meanwhile, funding bodies and universities were urged to support collaborative projects that test hybrid human-AI workflows and document best practices.
Conclusion and what to watch next
Specialists concluded that artificial intelligence in research can enhance research productivity and expand analytic reach if institutions pair it with critical thinking, methodological training and ethical safeguards. The next expected steps include pilot programs to build integrated research systems, updated academic guidelines on AI use and transdisciplinary studies on societal impact.
Readers should watch for institutional policy announcements, training initiatives and interdisciplinary projects over the coming months that aim to operationalize the balance between technological capability and human judgment.
