Copyright Protection for 3D Molecular Structures with Watermarking

2025-08-25

AI Summary

To address intellectual property protection challenges arising from AI-generated 3D molecular structures, this work proposes the first robust watermarking method tailored to molecular geometry. The method embeds imperceptible watermarks in the atomic-level feature space and leverages affine-invariant features to ensure strong robustness against rigid transformations—including rotation, translation, and scaling—while strictly preserving molecular topology, quantum chemical properties (>90% fidelity), and biological activity. Evaluated on the QM9 and GEOM-DRUG datasets using GeoBFN and GeoLDM generative models, the approach achieves >95% watermark detection accuracy. Molecular docking experiments confirm that watermarked molecules retain comparable binding affinity (−6.00 kcal/mol) and structural similarity to the native ligand (RMSD < 1.602 Å). This is the first systematic integration of digital watermarking into 3D molecular representation, establishing a verifiable and deployable paradigm for copyright protection in AI-driven drug discovery.

Technology Category

• Machine Learning: Quantum Machine Learning
• Computer Vision: 3D Computer Vision
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Security and Privacy: Data transparency and provenance
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Web Mining and Content Analysis: Web data provenance, reliability, and authenticity

Abstract

Artificial intelligence (AI) revolutionizes molecule generation in bioengineering and biological research, significantly accelerating discovery processes. However, this advancement introduces critical concerns regarding intellectual property protection. To address these challenges, we propose the first robust watermarking method designed for molecules, which utilizes atom-level features to preserve molecular integrity and invariant features to ensure robustness against affine transformations. Comprehensive experiments validate the effectiveness of our method using the datasets QM9 and GEOM-DRUG, and generative models GeoBFN and GeoLDM. We demonstrate the feasibility of embedding watermarks, maintaining basic properties higher than 90.00% while achieving watermark accuracy greater than 95.00%. Furthermore, downstream docking simulations reveal comparable performance between original and watermarked molecules, with binding affinities reaching -6.00 kcal/mol and root mean square deviations below 1.602 Å. These results confirm that our watermarking technique effectively safeguards molecular intellectual property without compromising scientific utility, enabling secure and responsible AI integration in molecular discovery and research applications.

Problem

• Protecting intellectual property of AI-generated molecular structures
• Ensuring watermark robustness against affine transformations
• Maintaining molecular integrity and scientific utility post-watermarking

Innovation

• Watermarking method using atom-level features
• Preserves molecular integrity and invariant features
• Maintains properties and accuracy above thresholds