| Abstract | n the era of ubiquitous digital communication, social media platforms have become primary channels
for information dissemination. However, this unprecedented connectivity has also facilitated the rapid
spread of misinformation, posing significant risks to public safety, democracy, and social cohesion.
Detecting and mitigating misinformation is a critical challenge for governments, media platforms, and
technology providers alike. In recent years, artificial intelligence (AI) models, including machine
learning (ML) and deep learning (DL) techniques, have shown promising capabilities in automatically
identifying and flagging false or misleading content. This survey paper systematically reviews the
current state of AI-driven misinformation detection on social media, examining conventional machine
learning models, state-of-the-art deep learning architectures, hybrid frameworks, and ensemble
techniques. The study also explores applications, publicly available datasets, evaluation metrics, and
highlights key challenges such as data imbalance, context dependency, and multilingual
misinformation. Finally, the paper proposes future research directions emphasizing explainability,
cross-platform detection, multimodal misinformation identification, and ethical considerations in AI
deployment for content moderation
|
|---|
