# An AI-driven conceptual framework for detecting fake news and deepfake content: a systematic review

## TL;DR

This paper reviews AI methods for detecting fake news and deepfakes, highlighting the need for integrated technical and governance solutions.

## Contribution

The paper introduces an integrated conceptual framework linking detection technologies, XAI, and governance mechanisms.

## Key findings

- There is a shift from CNNs to transformer- and CLIP-based architectures in deepfake detection.
- Persistent challenges include multimodal detection and cross-dataset generalization.
- The review emphasizes the need for robust benchmarks and interdisciplinary approaches.

## Abstract

The rapid advancement of generative artificial intelligence (AI) has enabled the creation of highly realistic synthetic media, commonly referred to as deepfakes, which are increasingly multimodal and difficult to detect. While these technologies offer creative and commercial potential, they also pose critical challenges related to misinformation, media trust, and societal harm. Despite the growing body of research, existing reviews remain fragmented, often separating technical detection advances from social and governance considerations. This study addresses this gap through a systematic review conducted in accordance with PRISMA guidelines across IEEE Xplore, Scopus, ACM Digital Library, and Web of Science. From an initial set of 120 database records, complemented by citation chaining, 34 studies published between 2014 and 2025 were included for analysis. Eighteen studies focused on deepfake generation and detection models, eight examined social and behavioural implications, and eight addressed ethical and regulatory frameworks. Thematic synthesis reveals a clear methodological shift from convolutional neural networks toward transformer- and CLIP-based architectures, alongside the emergence of large-scale benchmark datasets. However, persistent challenges remain in multimodal detection, cross-dataset generalization, explainability–robustness trade-offs, and the translation of governance principles into deployable systems. This review contributes an integrated conceptual framework that operationally connects detection technologies, explainable AI (XAI), and governance mechanisms through explicit feedback loops. Future research directions emphasize robust multimodal benchmarks, retrieval-augmented detection systems, and interdisciplinary approaches that align technical innovation with ethical and policy safeguards.