Tencent Zhuque Lab
Focused on real-world AI security offensive and defensive research, covering large model safety, AI agent security, and AI-generated content detection.
At a Glance
- AI developers
- Enterprise security teams
- Large Language Model providers
- Cybersecurity researchers
AI Tools by Tencent Zhuque Lab
(1)
AI-Infra-Guard (A.I.G)
AI Security Testing Platform
Discussions
No discussions yet
Be the first to start a discussion about Tencent Zhuque Lab
Latest News
AI-Infra-Guard (A.I.G) selected for Black Hat Arsenal
Release of SkillTrustBench with CUHK (Shenzhen)
Official release of the AI-Infra-Guard technical report
Publication of the specialized book 'AI Security: Technology and Practice'
Products & Services
A full-stack AI Red Teaming platform that integrates capabilities such as Agent Scan, Skills Scan, AI infra vulnerability scan, and LLM jailbreak evaluation.
A tool based on deep learning to identify AI-generated content (text, images, etc.) with high precision.
A systematic framework for analyzing security threats and attack paths in AI systems, modeled after the ATT&CK framework.
The industry's first evaluation platform for Large Language Models in the cybersecurity domain.
Market Position
A leading industrial AI security lab comparable to Google and OpenAI's internal Red Teams, distinguished by its focus on practical, tool-based offensive/defensive research.
Leadership
Founders
Yong Yang
Head of Tencent Security Platform Department; long-time security leader at Tencent.
Zheng Xing
Director of Tencent Zhuque Lab; previously held senior security research roles within Tencent.
Executive Team
Yong Yang
Head of Tencent Security Platform Department & Initiator of Zhuque Lab
Responsible for the overall security strategy at Tencent's Security Platform Department; initiated the founding of Zhuque Lab in 2019.
Zheng Xing
Director / Head of Tencent Zhuque Lab
Leads the lab's research on AI security, automated vulnerability updates, and benchmark alignment.
Board of Directors
Founding Story
Established in 2019 by the Tencent Security Platform Department to address emerging security risks in the AI era and protect Tencent's AI ecosystems.
Business Model
Revenue Model
Internal R&D funded by Tencent; contributions to open-source and global security standards.
Pricing Tiers
Available under Apache 2.0 license on GitHub.
Target Markets
- AI developers
- Enterprise security teams
- Large Language Model providers
- Cybersecurity researchers
- AI Red Teaming
- Model safety evaluation
- Detecting deepfakes and AI-generated misinformation
- Securing AI agent interactions
- Tencent
- DeepSeek
- Hugging Face
- NVIDIA