AI-DRIVEN attacks are accelerating security risk, with attackers using automation and large language models to rapid-fire through vulnerabilities after disclosure. The article cites Black Hat USA 2026 discussions and a Dark Reading reader poll showing that half of respondents view AI-driven attacks versus AI-powered defences in the SOC as the top theme, with automation and AI a distant second.
It notes that exploitation of high-profile flaws has begun within hours of public disclosure rather than days, aided by frontier AI models that help attackers identify and weaponise new weaknesses, as well as attackers’ ability to study patches and updates quickly. Notable references include the Hugging Face breach and ongoing discussions around NetScaler zero-days and Citrix, illustrating how AI can compress timelines and reduce the need for traditional technical disclosure to map updates and exploitation paths.
The piece highlights the pressure on defenders to keep pace. Omdia figures show 32% of organisations say AI-powered automated attacks have the greatest impact on offensive security strategies, while 25% point to faster exploitation as the trend most influencing their approach. For defence, analysts urge proactive use of AI to map permissions, trust relationships, and potential attack surfaces before exploitation occurs, and to consider controlled, staged auto-patching where mature processes exist.
The article also cites that 85% of organisations are increasing spending on offensive security, signalling a shift towards proactive stances to match machine speed. It emphasises that while auto-patching can help, it requires careful implementation and real-time monitoring to avoid unintended consequences. These findings come as organisations face the challenge of prioritising hundreds of CVEs and patches in a crowded threat landscape.