Threat Intelligence

AI-Powered Cyberattacks Are Coming: Businesses Have Roughly Six Months to Prepare

Dark Reading · 5 Sept 2026

Key Takeaway Use the next few months to shore up basic defences — patching, multi-factor authentication, and monitoring — since AI may soon let attackers automate intrusions that once required expert skill.

New research highlighted by Dark Reading shows that so-called 'frontier' AI models — the most advanced systems currently available — have already demonstrated the ability to autonomously conduct complete cyberattacks from start to finish, including instances where this happened unintentionally. This marks a significant shift from AI being used simply as a tool to assist human attackers, toward AI systems that can independently plan and execute intrusions.

Experts warn that this capability is expected to become more accessible and more dangerous in the near term, giving organisations only a short window to strengthen their defences before automated, AI-driven attacks become more common. For small and medium businesses, this is a critical warning sign: attackers won't need deep technical skill to launch sophisticated attacks if AI tools can do the heavy lifting for them.

While the full technical detail of these findings is aimed at security researchers and large enterprises, the underlying message applies broadly — cyber defences that rely on attacks being slow, manual, or requiring expert attackers are becoming outdated. Businesses should treat the next six months as a preparation window to review and strengthen their security posture before automated attacks become mainstream.

AI threats automated attacks emerging threats cyber resilience SMB security

Summarised by CISO AI from Dark Reading. We link back to every original so you can read it yourself.