The Journal of AI Production and Consumption immediately brings two thoughts to mind. On the one hand, AI production is about the building of AI tools and AI consumption deals with the use of AI tools. On the other hand, AI production and consumption describe the impacts of AI on the general production and consumption of goods and services.
No policies uploaded yet.
Manuscripts are initially checked by editors and a decision is made either to send them for review or to desk reject. The journal uses two anonymous peer reviewers, ordinarily subject experts. The review focuses on the academic honesty of authors, the integrity of manuscripts, the nurturing of young academic writers by seasoned professors, and the generation of AI literature for universities and research agencies.
We invite the submission of original research, reviews, and case studies that investigate the dynamic interplay between Artificial Intelligence and a diverse range of disciplines. We welcome contributions that explore both the application of AI to advance these fields and the critical examination of its societal impact.
Topics of interest include, but are not limited to:
Before submission, please ensure your manuscript aligns with the journal's aims and scope. The Journal of AI Production and Consumption (jAIpc) publishes original, high-quality research that critically examines artificial intelligence through a dual lens: as a product itself and as a transformative force on global systems.
We welcome studies that investigate:
The journal encourages interdisciplinary work and particularly welcomes submissions that contextualize these themes within the African experience, engaging directly with global dialogues. For a detailed description, please see our [link to Aims & Scope page].
jAIpc accepts the following types of submissions:
Manuscripts should generally include the following elements in order:
Ensure this main document is fully anonymized for peer review. Remove all author names, affiliations, and identifying information from the file properties.
All manuscripts must be submitted electronically through our online submission system: [Link to Online Submission Portal].
You will be required to:
jAIpc employs a double-anonymous peer review process. The Editor-in-Chief conducts an initial screening for scope and quality. Suitable manuscripts are then sent to at least two independent expert reviewers. The review process typically takes 8–12 weeks. Authors will be notified of the editorial decision: Accept, Minor Revisions, Major Revisions, or Reject.
Our copyright policy is designed to maximize the impact of your research. Authors retain copyright to their work and are encouraged to share it widely, which enhances its visibility and reach. In return, the journal retains the exclusive right to formally publish and manage the definitive Version of Record.
For all editorial and submission queries, please contact the editorial office at: jaipc.editorial@[journaldomain].org
Two times a year: May and October.
The personal information you provide (name, email address, and affiliation) is collected to facilitate the publication process for this journal. We are committed to protecting your data. We will never sell your personal information or share it with third parties for their marketing purposes. You have the right to access, correct, or request the deletion of your personal data at any time. For any questions about your data or this privacy policy, please contact us at [Insert Email Address of Journal Manager/Editor].
A non-refundable processing fee is required to support the publication of all accepted manuscripts. The fee amount is determined by the author's equity:
Please note:
All bank or transfer charges must be covered by the author.
We accept bank transfers or Western Union for international payments.
These fees are subject to change and may be periodically reviewed without prior notice.
Payment details will be provided with the formal acceptance letter.
Deep learning systems rely heavily on large and diverse datasets to achieve strong generalization performance. However, many real-world applications—particularly in developing regions—lack access to extensive image datasets, resulting in reduced model accuracy and increased overfitting. This research proposes a novel hybrid data augmentation strategy titled Flip-and-Hide, designed to support effective training of Convolutional Neural Networks (CNNs) in low-resource environments. The technique leverages two complementary transformations: spatial flipping and region hiding. Spatial flipping introduces geometric variability, while region hiding selectively masks random patches within an image to improve model robustness to occlusions and background noise. We evaluate the Flip-and-Hide technique on two popular benchmark datasets—Fashion-MNIST and CIFAR-10—using a baseline CNN architecture. The results are compared against three experimental conditions: (i) no augmentation, (ii) flipping only, and (iii) Hide-and-Seek region masking. Experimental findings show that the combined approach delivers more stable learning with reduced validation loss, while achieving an accuracy improvement of up to 4.8% compared to models trained with no augmentation. Statistical significance is confirmed through repeated trials across multiple epochs. This research demonstrates that simple yet strategically combined augmentations can significantly enhance image classification performance without additional dataset acquisition or computational overhead. The proposed technique offers a practical and scalable solution for institutions and researchers operating in data-constrained settings, particularly within developing countries where access to high-volume datasets remains limited.
Nagaye Wisdom, Yaw Afriyie