Microsoft: Hackers abusing AI at every stage of cyberattacks

Microsoft Threat Intelligence reports that adversaries are rapidly integrating generative and agentic AI into cyber operations—using models to craft phishing lures, fabricate realistic employee personas for remote IT worker schemes, generate or debug malware, provision infrastructure, and evade defenses. The report names North Korean-linked groups (Jasper Sleet, Coral Sleet) and highlights techniques such as jailbreaking language models and AI-assisted scaling of campaigns, warning defenders to focus on identity protections, abnormal credential use detection, and AI system security.