We are all certainly familiar with the current potential and capabilities of Artificial Intelligence (AI). From solving mathematical problems and creating designs to editing photos, videos, and managing big data, AI has demonstrated adequate proficiency in executing these tasks.

AI learning methods are categorized into two primary types: Machine Learning and Deep Learning. Machine Learning (ML) is a branch of AI focused on mapping machines or programs to recognize, understand, and respond systematically. Generally, ML utilizes less complex data and requires a relatively short time to implement.

Furthermore, Deep Learning (DL) is a derivative of Machine Learning capable of generating content rather than merely delivering it. Unlike ML, Deep Learning utilizes "Artificial Neural Networks" to process larger datasets with significantly higher levels of complexity. Overall, DL excels at deep analysis while simultaneously generating unique patterns. Common implementations of DL include autonomous vehicles, facial recognition, and more.

With all its sophistication, will AI become the ultimate answer to all human inquiries and render research—as a source of authentic and valuable information—obsolete? In short, no. Much like any system that evolves according to the user's competency, AI, Machine Learning, and Deep Learning are limited by one factor: consistency.

To address the challenges of increasingly aggressive automation, a seminar titled "Preparation of Scopus-Indexed Journals and the Utilization of ChatGPT" was held. Located at the Aula Merah Putih, UNHAN RI, the event featured several expert speakers, including Mr. Caturida Meiwanto Doktoralina, a lecturer from the Accounting Study Program at Dian Nusantara University (UNDIRA), who also possesses a professional background at the National Resilience Institute of the Republic of Indonesia (Lemhannas RI).

According to him, total reliance on technology and automation will undermine the quality of research if it is not balanced with mastery, literacy, and a deep understanding of the subject matter. Mr. Caturida emphasized that the growth of a critical mindset cannot be freely replaced by AI.

"AI cannot be fully trusted in the scientific research process. Its use must be limited and controlled, particularly in the Methods, Results and Discussion, and Conclusion sections. Scientific substance, theoretical arguments, and the interpretation of findings remain the full responsibility of the author," he stated.

From an educator's perspective, Mr. Caturida also highlighted that while technology has simplified basic reasoning processes, the role of lecturers as guardians of academic quality remains essential. Amidst an influx of information prone to misconceptions, human intervention is necessary to rectify confirmation bias or subjective misunderstandings among students.

"Even when technology is utilized, lecturers must maintain academic control. Scientific writing has specific conventions, an academic tone, and scholarly standards that differ from popular communication. The propriety, validity, and reliability of research results are non-negotiable. Furthermore, this pertains to the Regulation of the Minister of Education, Culture, Research, and Technology Number 39 of 2021 concerning Academic Integrity in Producing Scientific Works," he added.

This digital literacy effort by Mr. Caturida Meiwanto Doktoralina is a tangible manifestation of UNDIRA's commitment to cultivating an excellent generation. Grounded in the core values of Visionary, Integrity, and Professionalism (VIP), UNDIRA recognizes that technological advancements like AI and Deep Learning are not merely fleeting trends, but future challenges that must be approached with wisdom.

UNDIRA does not merely focus on producing "one-time solutions"; rather, it consistently prioritizes output, quality, and educational and research standards that are scientifically accountable and provide sustainable benefits to the wider community.

Source of References: UNHAN RI: AI Bukan Pengganti Nalar Akademik dalam Publikasi Ilmiah