Prompt injection attacks hit over 90 firms in 2025 as AI security risks escalate
CrowdStrike's 2026 Global Threat Report reveals that threat actors targeted more than 90 organizations with prompt injection attacks in 2025, stealing credentials and cryptocurrency. The report documents an 89% year-over-year surge in AI-enabled adversary operations, with prompt injection now functioning as malware by exploiting enterprise AI design flaws in agents, RAG pipelines, and model routers.