Anthropic has published research on AI-enabled cyber threats after examining 832 accounts banned for malicious cyber activity between March 2025 and March 2026.
The company mapped the activity to MITRE ATT&CK and found that AI is not only being used for early-stage preparation. Threat actors are increasingly using AI in later and more complex parts of cyber operations, including activity after a system has already been compromised.
The report is a reminder that AI security is becoming a business issue, not just a technical issue. Any company using agents, automation, code tools, or AI-connected systems needs stronger monitoring, access control, and human review around high-risk workflows.
Source: Anthropic, published June 3, 2026. Read the source update.