Regulatory Compliance Automation in MCP Deployments
Agents need runtime governance because MCP shifts compliance decisions from code to behavior.
Priya Subramaniam
Senior Correspondent, AI Governance
Priya Subramaniam
Priya Subramaniam spent eight years covering data protection and enterprise risk for a consortium of European regulatory publications before pivoting to AI policy in 2019. She focuses on the intersection of compliance frameworks and autonomous systems, with particular attention to how organizations translate regulatory mandates into operational controls.
12 stories
NIS2 Directive Implications for AI-Driven Enterprise Operations
EU regulators are enforcing AI governance rules that NIS2 never explicitly mention.
Priya Subramaniam
HIPAA Considerations for AI Agents Accessing Health Data via MCP
Build HIPAA controls into the MCP server itself, not as an afterthought downstream.
Priya Subramaniam
SOC 2 Controls for MCP-Enabled AI Environments
Autonomous agents break SOC 2's core assumptions about control and accountability.
Priya Subramaniam
GDPR Compliance When AI Agents Access Personal Data via MCP
Thousands of exposed AI servers put enterprises at risk of major GDPR violations.
Priya Subramaniam
EU AI Act Requirements for Agentic AI Systems
Enterprises must map agentic AI to existing Act rules without a special regulatory category.
Priya Subramaniam
Prompt Injection Attacks Targeting MCP Tool Calls
Attackers exploit LLMs' inability to distinguish instructions from data in tool descriptions.
Priya Subramaniam
MCP Tool Schema Design for Least Privilege
Three fields in your tool schema determine whether AI agents can run wild or stay locked down.
Priya Subramaniam
Change Management for Enterprise AI Governance Programs
Most companies have AI governance policies on paper but fail to actually operate them day to day.
Priya Subramaniam
Agentic AI Incident Response Playbooks
Governance frameworks that let agentic AI respond faster without losing audit trails.
Priya Subramaniam
Governance Frameworks for Autonomous AI Agents
How to build identity, access control, and observability before agents go live.
Priya Subramaniam
Shadow AI and Ungoverned MCP Adoption in Enterprises
Most enterprises can't see the AI tools their workers use, let alone control them.
Priya Subramaniam