Viewing AI Studies from an IJTech Perspective

Over the past decade, artificial intelligence (AI), machine learning, and neural-network-based methods have moved from the margins of engineering research into the mainstream. This trajectory is clearly visible in the publication statistics of International Journal of Technology (IJTech). Using our internal data from 2016–2025, this editorial note reflects on the rise of AI-themed submissions and publications in IJTech and discusses how their citation performance compares, in broad terms, with the journal’s generic themes.

Our dataset identifies articles whose title, abstract, or keywords explicitly include “artificial intelligence,” “machine learning,” “neural network,” or closely related terms. From 2016 to 2025, IJTech published 90 such AI-themed papers, which together received 712 citations—an average of 7.9 citations per article over the period. This already indicates that AI-related work is not only numerous but also visible within our citation landscape.

When we look at the share of AI papers in the journal’s overall output, the growth is striking. Between 2017 and 2025, IJTech published 1,254 papers across all themes. In the same years, 86 of these were AI-themed, meaning that roughly 7\% of all IJTech publications since 2017 explicitly use AI, machine learning, or neural-network approaches. In the early years, AI was truly a niche: in 2017–2019, AI contributions accounted for only around 2–3\% of all published papers per year. By 2022, this share had increased to about 8.5\%, and in 2023 AI papers represented more than 13\% of that year’s output. In 2024, they remained at about 10\% of all publications.

The data for 2025 are particularly notable, although they must be interpreted cautiously because the year is still incomplete. In the current snapshot, only 26 papers are recorded as published, and 20 of these are AI-themed. In other words, nearly 77\% of the 2025 publications so far fall into the AI/machine learning/neural-network category. Even allowing for additional non-AI papers that will be added as the year closes, it is clear that AI has become one of the dominant themes in our recent issues.

Citation patterns within this AI subset also tell an important story. Early AI papers, published when the topic was still emerging in our community, attracted relatively high citation densities: in 2016–2019 the average ranged from about 7 to 18 citations per article, and in 2020 it peaked at more than 28 citations per article. These papers often introduced AI tools into areas where they were still novel—such as predictive maintenance, process optimization, energy systems, and transport modeling—so each contribution had strong visibility.

As AI became more common and the number of AI-themed papers increased, the average citations per AI article began to moderate: around 8 citations per paper in 2021, 9 in 2022, 6 in 2023, and lower values so far for 2024–2025 (partly because these papers are still very recent and have not had time to accumulate citations). Even so, across the decade, AI contributions as a group maintain a healthy citation/article ratio that is broadly comparable to, and in several years appears to be above, the typical performance of more generic themes in the journal.

For IJTech, these patterns have several implications. First, AI is no longer a special topic but a cross-cutting methodology that now appears in roughly one out of every ten papers we publish, and in some recent issues, a much higher fraction. Second, the most visible AI papers are not those that merely apply fashionable algorithms, but those that combine robust data, careful experimental or numerical design, and genuine technological insight. Finally, as AI becomes mainstream, our editorial emphasis will shift increasingly from “Is this AI?” to “Does this AI-based work advance technology and engineering in a meaningful, rigorous, and responsible way?”.