Detecting AI-generated Artwork

2025-04-09

AI Summary

Amidst the proliferation of AI-generated art, detecting forgeries across diverse, high-complexity artistic styles—particularly Baroque, Cubism, and Expressionism—remains a critical yet underexplored challenge. Method: We construct and systematically evaluate multiple machine learning and deep learning models to distinguish AI-generated images from human-created artworks across these three styles. We introduce the first benchmark framework for multi-style AI art forensics, supporting both binary classification (AI vs. human) and fine-grained six-class classification (3 styles × 2 author types). Contribution/Results: Convolutional Neural Networks (CNNs) achieve the highest robustness in binary classification (97.58% accuracy) and strong performance in six-class classification (82.08% accuracy), substantially outperforming human observers’ inconsistent judgments. This work is the first to empirically demonstrate the efficacy of deep visual features for cross-style AI art detection, establishing a reproducible methodological foundation and empirical evidence base for assessing artistic authenticity.

Technology Category

• Computer Vision: Generative Adversarial Networks (GANs) for Vision
• Humans and AI: Other Foundations of Human Computation & AI
• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification

Application Category

• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI
• Web Mining and Content Analysis: Web data quality in the era of algorithmically-generated content
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

The high efficiency and quality of artwork generated by Artificial Intelligence (AI) has created new concerns and challenges for human artists. In particular, recent improvements in generative AI have made it difficult for people to distinguish between human-generated and AI-generated art. In this research, we consider the potential utility of various types of Machine Learning (ML) and Deep Learning (DL) models in distinguishing AI-generated artwork from human-generated artwork. We focus on three challenging artistic styles, namely, baroque, cubism, and expressionism. The learning models we test are Logistic Regression (LR), Support Vector Machine (SVM), Multilayer Perceptron (MLP), and Convolutional Neural Network (CNN). Our best experimental results yield a multiclass accuracy of 0.8208 over six classes, and an impressive accuracy of 0.9758 for the binary classification problem of distinguishing AI-generated from human-generated art.

Problem

• Distinguish AI-generated from human-generated artwork
• Evaluate ML/DL models for art classification
• Focus on baroque, cubism, expressionism styles

Innovation

• Uses Machine Learning and Deep Learning models
• Tests Logistic Regression, SVM, MLP, CNN
• Achieves high accuracy in art classification