AI驱动网络攻击进入“分钟级”，企业安全逻辑亟待重构

Core Insights - The average time required for a successful cyber attack has drastically decreased from 9 days in 2021 to as little as 25 minutes by 2025, highlighting the urgent need for companies to rethink their security strategies [1][3][4] - The rise of AI in both offensive and defensive roles is reshaping the cybersecurity landscape, necessitating a shift from traditional passive defense mechanisms to more proactive and integrated security architectures [2][4][7] Group 1: Cyber Attack Trends - AI-related cyber and data security incidents have significantly increased, with the time to execute successful attacks reducing from 9 days in 2021 to 2 days in 2023, and potentially to 25 minutes by 2025 [3][4] - The speed of attacks has improved over 100 times in three years, creating a growing imbalance between attackers and defenders [4] Group 2: Security Challenges - Traditional security models, which focus on network boundaries, are becoming obsolete due to the expansion of attack surfaces from remote work and AI integration [7] - Many enterprises utilize an average of 83 different security tools from 29 vendors, leading to fragmented data and a false sense of security [7] Group 3: Strategic Recommendations - Companies should transition to a platform-based, integrated security architecture to enhance efficiency and reduce the number of tools from 83 to fewer than 10 [8] - Implementing security measures early in the AI development lifecycle is crucial, including selecting trusted base models and filtering sensitive training data [8] - The future of cybersecurity will hinge on capabilities such as global visibility, real-time response, cross-domain defense, and continuous evolution to address unknown threats [8]