AI Governance & Compliance

EU AI Act, ISO 42001, the LOOP™ methodology: our analyses on AI agent governance, written for regulated enterprises and their compliance and risk teams.

Browse by category
All articlesAdopting AI in the EnterpriseAI Agents in ProductionAI Governance & ComplianceClaude & Anthropic EcosystemLearning Claude
AI Governance & ComplianceMCP
September 30, 2026·13 min
MCP audit in production: risks and a checklist for your agents’ servers
Model Context Protocol in production: authorisation, provenance, isolation and logging of MCP servers, plus the audit checklist to run server by server in your stack.
Read the article
AI Governance & ComplianceREDTEAM
September 30, 2026·13 min
Red teaming AI agents: how to test a Claude agent before and after go-live
Prompt injection, data exfiltration, guardrail bypasses: the method for testing a Claude agent before deployment, then on an ongoing basis once it is in production.
Read the article
AI Governance & ComplianceSECURE
September 30, 2026·13 min
Securing a Claude agent in production: the complete guide
Permissions, secrets, logging, human oversight: the guide to securing a Claude agent in production, with LOOP™ trust zones applied to security risk in the enterprise.
Read the article
AI Governance & ComplianceART. 50
August 3, 2026·17 min
The Digital Omnibus delay explained: what moved in the AI Act, what did not
The Digital Omnibus package pushed high-risk obligations to December 2027 and left Article 50 applicable since August 2026. What moved, what holds, where to start.
Read the article
AI Governance & ComplianceCAIO
July 20, 2026·15 min
Chief AI Officer: what the role does and whether you need one
Orange has appointed a new Chief AI Officer. What the role does, what Orange's mandate reveals, and how to decide whether your company should appoint one too.
Read the article
AI Governance & ComplianceLOOP
April 1, 2026·16 min
LOOP™: the human-centered AI governance protocol
LOOP™ sets how much autonomy an AI agent gets: 4 trust zones, 3 escalation levels and a living registry. How it works, who owns it and how it maps to the EU AI Act.
Read the article
AI Governance & Compliance4Z
March 6, 2026·17 min
The 4 trust zones: classifying your AI agents
Green, orange, red or black: how the LOOP™ method classifies each action of your AI agents before production, with its criteria and its common pitfalls.
Read the article