All Writing

The road ahead: What evidence would justify more autonomy?

Before removing an approval step, ask what evidence would justify the change, what the trial cannot show, and what would reverse the decision.

The Amplification Playbook

A proposed trial for redesigning one workflow around AI, with explicit baselines, worker input, learning checks, and a decision to stop.

Incident resolution with AI: A hypothetical reference design

A fictional incident workflow follows an AI-assisted rollback from alert to verified recovery, with explicit authority, handoff, and failure constraints.

The New Skill Premium

Job postings show which skills employers ask for, not what caused a wage premium. Build a development plan without mistaking either for a guarantee.

From vibe coding to verified: Practising independent diagnosis

A proposal for practising independent diagnosis in AI-assisted engineering, with synthetic exercises and tests of whether the learning transfers.

What sleeper agents mean for production AI systems

Sleeper-agent experiments show why cleaner evaluations do not prove harmful behaviour is gone, and why production checks need independent evidence.

Amplifying Creativity

AI suggestions can improve an individual story while making a collection of stories more alike. Quality and variety need separate tests.

Designing Decision Boundaries for AI Autonomy

How to turn autonomy policy into testable action conditions, using a hypothetical booking service to examine consent, reversibility, retries, and the cost of delay.

Intelligence Amplification in Code

AI coding tools can accelerate a bounded task and slow experienced maintainers. Measure the work you actually need to finish.

Five Ways to Question an AI Agent's Success

A proposed review framework for checking an agent's diagnosis, scope, lasting outcome, downstream effects, and context freshness beyond a successful tool call.

AI Governance Needs Shared Rules and Different Controls

How to keep shared AI governance rules while tailoring controls to each use, with explicit ownership, testable evidence, and reassessment when permissions change.

The Cognitive Offloading Trap

Better assisted work and better learning are different outcomes. A classroom experiment shows why the distinction matters.

When Human Oversight Pays for Itself

A practical cost model for human review, with a hypothetical payback calculation, sensitivity checks, and cases where oversight does not break even.

The Great Equalizer

AI can narrow performance gaps on a task. That does not mean it gives novices an expert's knowledge, pay, or independence.

Your AI Agent's Skills Might Be Malware

What the ToxicSkills audit found, what its numbers do not prove, and how to limit the damage from a compromised agent skill.

Graduated Automation: How to Scale AI Without Losing Control

How a procedure earns permission to act: scoped approvals, independent checks, and withdrawing autonomy when conditions change.

The Centaur Advantage

Human-AI routing only helps if the router can identify a useful difference in ability. A chess experiment shows both the opportunity and the difficulty.

The Hidden Debt in AI-Generated Code

What public studies reveal about persistent code-quality issues, the limits of AI-versus-human comparisons, and reviewing architectural change.

The Amplification Thesis

AI can improve human work without making human-plus-AI the best option. Design for amplification, then test it against the alternatives.

AI and critical thinking: What the research actually shows

What an EEG experiment, a knowledge-worker survey, and a perspective paper tell us about AI, critical thinking, and meaningful human oversight.

AI changes the work. Who gets the gains?

Automating a task can help a worker, replace their work, or create demand elsewhere. The outcome depends on how jobs and institutions change.

Why Human-in-the-Loop is Not Optional for Production AI Agents

The six-pillar argument for why HITL is not a transitional compromise but a structural requirement of any production AI system.

Decentralized Granular Access Control for Agentic AI Systems in Critical Infrastructure

A decentralized, multi-layered access control architecture for agentic AI in critical cloud infrastructure.

The Case for Human-in-the-Loop in Production AI Agent Systems: A Six-Pillar Framework

53 research papers argue HITL is a structural requirement of any production AI system.

From Reactive to Autonomous: Evolution of AI Operations in Cloud Network Infrastructure

Five-generation maturity model from manual troubleshooting to autonomous incident resolution.

Autonomous Incident Resolution at Hyperscale: An Agentic AI Architecture for Network Operations

Multi-agent orchestration framework for autonomous network incident resolution with safety guarantees.

MCP vs CLI: A Comparative Analysis of AI Agent Tooling Interfaces

CLI and MCP are complementary transport layers. Evaluated across 14 dimensions with benchmarks.

Zero Trust in Azure: Strategies for Securing Cloud Infrastructure

Seven practical strategies covering NSGs, Private Links, Azure Policies, AVNM, and shift-left practices.