The post tests four controls for safely stopping autonomous AI agents: external kill switches and iteration caps, spend ceilings, human approved pauses, and pattern breakers for persistent loops. It recommends fail closed, fleet wide controls, immutable…
https://t.co/hVUd72sWEJ
This blog discusses the necessity for organisations to implement effective AI kill switches as they deploy autonomous systems. It contrasts the misleading term "kill switch" with the reality of operational safety mechanisms.
https://t.co/BTZeQU2IUy
This post concludes a three-part series on scaling agentic AI. It emphasises the challenges of governing a vast ecosystem of autonomous agents, highlighting the inadequacy of traditional governance methods.
https://t.co/bcUYqDQ0IO
This is a companion piece to What Should a Board Be Confident About Before Scaling AI Agents? which sets out the six questions every board should answer before approving agentic AI at scale.
https://t.co/qNDGPdbnDR
Organisations are urged to scale AI agents, but must first answer six critical governance questions. Issues of trust, accountability, and architecture are highlighted as major barriers.
https://t.co/EQmGU2AQmm
Enterprise architects often fail to gain stakeholder support not due to poor technical decisions, but because their communication does not effectively convey the business implications of those decisions.
https://t.co/1ZQJ5OrGgl
Architecture Decision Records (ADRs) are becoming essential for AI systems. They capture the architecturally significant choices that shape model behaviour, risk posture and compliance.
https://t.co/VcZ7EyMe5l
Ensuring supply chain integrity is becoming essential for trustworthy AI. With risks ranging from data and model poisoning to recent high profile cybersecurity breaches, organisations need clearer visibility into what their AI systems are built from.
https://t.co/Um0I6NVZmG
AI systems now act autonomously, pushing governance from policy documents to real‑time control. The OpenAI and Anthropic breaches showed how exposed organisations are when authority isn’t enforced at execution.
https://t.co/jfUe5jlSfQ
Two comments on the Agent Registry post identified the same architectural truth: visibility alone is not governance. This post acknowledges the gap and introduces a governance tetrad, the four sided structure that makes a registry load-bearing.
https://t.co/2SaA2sl6u0
The emergence of AI agent sprawl poses significant governance challenges, leading to compliance risks and forensic investigation difficulties. Organisations struggle to maintain visibility over numerous agents, often resulting in unsanctioned systems.
https://t.co/UCmvXbAT3O
AI Forensic Readiness: Evidence Architecture for the Agentic Enterprise
When AI can fabricate evidence indistinguishable from reality, the “Some AI Did It” defence becomes every investigator’s nightmare.
https://t.co/wdt7YcKxMS
Why Shadow AI is the Defining Enterprise Risk
Shadow AI usage is a growing enterprise risk, as employees increasingly use unsanctioned AI tools without IT oversight.
Latest blog
https://t.co/HSJSP9IAJK
EA preserves strategic flexibility by enabling deliberate commitments that shape future options. Through modularity, loose coupling and data portability, it avoids short‑term rigidity and supports resilience amid regulatory and market change.
https://t.co/uYR43qR9hL
Effective presenting hinges on engaging the audience through connection and inspiration. To enhance delivery, focus on smooth, flowing expression rather than rhythm alone. Overcome rigid bearing with relaxation techniques and practice emotional immersion.
https://t.co/m5xzhbHDr4
Today, I celebrate reaching 150 hours of volunteering with STEM Ambassadors, earning a Bronze badge. This journey has been fulfilling, helping young people explore STEM fields.
https://t.co/eOjBbJYNbo
Modern organisations often rely on simple diagrams for architecture, which can lead to inconsistent interpretations and governance challenges. Adopting structured modeling languages enhances clarity, communication and strategic alignment.
https://t.co/0OTEPqaxPZ
As of 2026, organisations face a two year countdown to first key milestone outlined by the NCSC for migrating to post quantum cryptography (PQC) due to quantum threats. Enterprise Architecture (EA) is crucial for planning, standardising & managing.
https://t.co/zcZsz7jX0D