Artificial Intelligence–Based Detection of Deepfake Media on Social Platforms: Evaluating Machine Learning Models for Identifying Manipulated Video Content

Ratan Singh

Artificial Intelligence–Based Detection of Deepfake Media on Social Platforms: Evaluating Machine Learning Models for Identifying Manipulated Video Content

Author(s)	Mr. Ratan Singh
Country	India
Abstract	Deepfakes are synthetic or manipulated audio-visual media produced through deep learning techniques that can depict people saying or doing things that did not occur. Their increasing realism has intensified concerns about misinformation, fraud, privacy, political communication, and the evidentiary reliability of digital media. This paper presents a critical narrative review of deepfake detection research, comparing traditional forensic methods, classical machine learning approaches, and contemporary deep learning systems. The review examines spatial detectors based on image artefacts, temporal models that analyse frame-to-frame inconsistency, physiological approaches that use biological signals, and multimodal systems that combine visual and audio evidence. The literature indicates that deep learning methods generally outperform handcrafted forensic techniques on benchmark datasets because they learn complex features directly from data. However, benchmark superiority does not translate automatically into dependable real-world performance. Cross-dataset testing reveals substantial losses in accuracy when detectors encounter unfamiliar manipulation methods, compression, low resolution, demographic variation, or adversarial perturbations. The central conclusion is therefore conditional: deep learning provides the strongest current detection capability, but no single architecture is sufficiently general, explainable, efficient, and robust for universal deployment. Future progress requires diverse and continually updated datasets, cross-dataset evaluation, multimodal fusion, uncertainty-aware decisions, adversarial testing, provenance systems, and governance frameworks that protect privacy and due process. Deepfake detection should be treated as one component of a broader media-authenticity ecosystem rather than as a complete solution to synthetic-media harms.
Keywords	deepfake detection, synthetic media, digital forensics, convolutional neural networks, multimodal detection, generalisation, media authenticity
Published In	Volume 8, Issue 4, July-August 2026
Published On	2026-08-10

View / Download PDF File

About IJFMR Fees & Payment Current Issue Publication Archive	Submit Research Paper Track Submission Status Publication Guidelines Publication Ethics Peer Review & Plagiarism	Join as a Reviewer Editors & Reviewers Reviewer Referral Program Get Reviewer Membership Certi.	Website/Journal Policies Usage Policy Content Policies Privacy Policy

Contact Us		+91-9687-828-838	editor@ijfmr.com

International Journal For Multidisciplinary Research

A Widely Indexed Open Access Peer Reviewed Multidisciplinary Bi-monthly Scholarly International Journal

Artificial Intelligence–Based Detection of Deepfake Media on Social Platforms: Evaluating Machine Learning Models for Identifying Manipulated Video Content

Share this