Research on content security of generative AI video empowered by internet of computing

doi:10.12267/j.issn.2096-5931.2026.02.010

Abstract

Abstract:

Generative artificial intelligence （AI） video technology is driving innovation in content creation，but also brings some challenges. Focusing on three typical tasks—Text-to-Video （T2V），Image-to-Video （I2V），and AI-powered video editing—this study analyzes their distinct risks and systematically sorts out general security technologies covering the entire lifecycle of pre-generation，in-generation，and post-generation. It discusses precise adaptation strategies for different tasks and the role of the internet of computing. By examining industry practices，it identifies key governance shortcomings，including the gap in security capabilities，fragmented identification standards，and ambiguous responsibility definition. This paper provides support for building a targeted AIGC video security system.

Key words: internet of computing, video generation, generative AI, video content security

CLC Number:

TP391.4

YANG Yihe, MA Fei, LI Zhe, DONG Xiaohui. Research on content security of generative AI video empowered by internet of computing[J]. Information and Communications Technology and Policy, 2026, 52(2): 69-77.

Add to citation manager EndNote|Ris|BibTeX

URL:

http://ictp.caict.ac.cn/EN/10.12267/j.issn.2096-5931.2026.02.010

http://ictp.caict.ac.cn/EN/Y2026/V52/I2/69

Figures/Tables 2

References 19

[1]	XIE F, ZENG D, SHEN Q, et al. A comprehensive survey on text-to-video generation[J]. Chinese Journal of Electronics, 2025, 34(4): 1009-1036. doi: 10.23919/cje.2024.00.151 URL
[2]	REN W, YANG H, ZHANG G, et al. ConsistI2V: enhancing visual consistency for image-to-video generation[J]. arXiv Preprint, arXiv:2402.04324, 2024.
[3]	CEYLAN D, HUANG C H P, MITRA N J. Pix2Video: video editing using image diffusion[C]// Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 2023: 23206-23217.
[4]	LIU C, YU H. AI-empowered persuasive video generation: a survey[J]. ACM Computing Surveys, 2023, 55(13s): 285.
[5]	OpenAI. Video generation models as world simulators[R], 2026.
[6]	SINGER U, POLYAK A, HAYES T, et al. Make-a-video: text-to-video generation without text-video data[J]. arXiv Preprint, arXiv:2209.14792, 2022.
[7]	MIAO Y B, ZHU Y F, YU L J, et al. T2VSafetyBench: evaluating the safety of text-to-video generative models[J]. Advances in Neural Information Processing Systems, 2024, 37: 63858-63872.
[8]	YOON J, YU S, PATIL V, et al. Safree: training-free and adaptive guard for safe text-to-image and video generation[J]. arXiv Preprint, arXiv:2410.12761, 2024.
[9]	BLATTMANN A, DOCKHORN T, KULAL S, et al. Stable video diffusion: scaling latent video diffusion models to large datasets[J]. arXiv Preprint, arXiv:2311. 15127, 2023.
[10]	栗蔚, 张博圣, 孙松林, 等. 算力互联网架构:基于熵平衡支持算力资源跨域互联的下一代网络架构[J]. 通信学报, 2025, 46(9):1-16.
[11]	LI J, LI D, SAVARESE S, et al. Blip-2: bootstrapping language-image pre-training with frozen image encoders and large language models[C]// International Conference on Machine Learning. PMLR, 2023: 19730-19742.
[12]	TANG T, WU Y, WU Y, et al. Videomoderator: a risk-aware framework for multimodal video moderation in e-commerce[J]. IEEE Transactions on Visualization and Computer Graphics, 2021, 28(1): 846-856. doi: 10.1109/TVCG.2021.3114781 URL
[13]	YUAN Y, SONG J, IQBAL U, et al. Physdiff: physics-guided human motion diffusion model[C]// Proceedings of the IEEE/CVF International Conference on Computer Vision. 2023: 16010-16021.
[14]	HUANG Y, CHEN J, ZHENG Q, et al. Video signature: in-generation watermarking for latent video diffusion models[J]. arXiv Preprint, arXiv:2506.00652, 2024.
[15]	WANG Q, YU G, SAI Y, et al. Is your AI truly yours? Leveraging blockchain for copyrights, provenance, and lineage[J]. arXiv Preprint, arXiv:2404.06077, 2025.
[16]	OpenAI. Advancing red teaming with people and AI[R], 2024.
[17]	MOUSTAFA M. Applying deep learning to classify pornographic images and videos[J]. arXiv Preprint, arXiv:1511.08899, 2015.
[18]	SIMMONS J C, WINOFRAD J M. Interoperable provenance authentication of broadcast media using open standards-based metadata, watermarking and cryptography[J]. arXiv Preprint, arXiv:2405.12336, 2024.
[19]	WANG Q, LI C, LUO Y, et al. Detecting adversarial data using perturbation forgery[C]// Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Piscataway: IEEE, 2025: 13917-13926.

	T2V	I2V	AI视频编辑
生成前	文本提示过滤利用OCR/ASR反馈的闭环式提示过滤	多模态风险评估图像哈希比对	编辑指令语义审查原始视频合规性检查
生成中	安全对齐微调轻量级安全适配器潜空间语义一致性约束	潜空间动态约束（运动合理性、生物力学先验等）	弱全局安全约束敏感对象修改限制
生成后	绑定完整生成上下文的水印	绑定输入图像及生成上下文的水印语义放大效应检测	绑定编辑操作日志的水印生物痕迹建模多模态伪造检测生成器指纹溯源

	T2V	I2V	AI视频编辑
生成前	文本提示过滤利用OCR/ASR反馈的闭环式提示过滤	多模态风险评估图像哈希比对	编辑指令语义审查原始视频合规性检查
生成中	安全对齐微调轻量级安全适配器潜空间语义一致性约束	潜空间动态约束（运动合理性、生物力学先验等）	弱全局安全约束敏感对象修改限制
生成后	绑定完整生成上下文的水印	绑定输入图像及生成上下文的水印语义放大效应检测	绑定编辑操作日志的水印生物痕迹建模多模态伪造检测生成器指纹溯源