Experts warn hackers are using AI chatbots to write malware using natural language (www.techradar.com)

Huntress has revealed a concerning trend where low-skilled cybercriminals are leveraging generative AI to create custom malware, highlighted by their analysis of a malware sample called "Untitled1.ps1." This bespoke Active Directory (AD) enumeration tool demonstrates how AI can lower the barriers to cybercrime, allowing unsophisticated attackers to generate unique, evasive payloads that traditional security measures, reliant on static signatures, often fail to detect. This shift to AI-generated code, which diverges from previous off-the-shelf options, poses a significant challenge for cybersecurity defenders. The malware's ability to map AD environments before employing tools like s5cmd for rapid data exfiltration signifies a new level of threat sophistication. Researchers emphasize that while the generated code may appear messy and over-engineered, its potential for executing devastating data breaches requires defenders to pivot from rigid, signature-based methods to more flexible behavioral analytics strategies. By focusing on the underlying actions of an attack, defenders can better identify and mitigate these emerging threats driven by AI, which underscore the urgency for enhanced cybersecurity measures in a landscape increasingly influenced by AI capabilities.