Archive

"Enough to Reconstruct, Never Enough to Leak": The HIPAA...

Enough to reconstruct, never enough to leak. The audit event schema problem under §164.312(b), and how to solve it without conflating the audit sink with the PHI sink.

Emergence and Complexity Classes — How Simple Rules Behave

Run enough cellular automata and a pattern in the patterns appears: their behavior falls into a few distinct classes, from dull to chaotic to the fascinating edge between. Wolfram's classification of this behavior, and the deep idea that the most interesting computation lives at the "edge of chaos," are among the most thought-provoking results in complexity science. This post is about how cellular automata behave, why complexity concentrates at a boundary, and the surprising unpredictability lurking in deterministic rules.

CA behavior falls into Wolfram's four classes — homogeneous, periodic, chaotic, and complex — with complexity and computation concentrating at class 4, the 'edge of chaos' between sterile order and sterile chaos. Plus computational irreducibility: for complex systems there's no shortcut, so simple deterministic rules can be fundamentally unpredictable.

The Future and Responsible AI in Energy

AI's potential to help operate a clean, complex grid is enormous — and precisely because the stakes are so high, this is exactly the domain where AI must be applied most carefully. A wrong forecast is one thing; an AI decision that destabilizes critical infrastructure serving millions is another entirely. The future of AI in energy is not "hand the grid to an AI" but something more disciplined and more valuable: AI as trustworthy decision-support that helps human operators run a decarbonized grid safely. This closing post is about that future, and the responsibility it demands.

AI's potential to help operate a clean, complex grid is enormous — and precisely because the stakes are so high, this is exactly where AI must be applied most carefully. The future is not 'hand the grid to an AI' but something more disciplined: AI as trustworthy decision-support that helps human operators run a decarbonized grid safely.

providers/azure · Azure AI Project

Point the same agent at an Azure AI Foundry project endpoint using the project Responses API mode.

Run an agent against an Azure AI Foundry project: foundryprovider.NewAgent plus ModelDeployment selects project Responses API mode from the project endpoint.

The Fallback Is the Contract: Reliability Patterns for...

Every LLM-backed agent in this platform has a deterministic rule-based fallback. The case always finalises. The fallback isn't a workaround — it's the contract.

Multi-Agent Systems in 5 Interfaces: A Microsoft Agent...

Five interfaces hold the whole platform together. The 30-line orchestrator closure that makes the rest of the architecture testable, auditable, and safe to evolve.

Conway's Game of Life — Four Rules, Infinite Behavior

The most famous cellular automaton in the world has four rules, a 2D grid of cells that are alive or dead, and no players at all. Conway's Game of Life is "played" by watching it unfold — and from its four trivial rules emerge gliders that travel, oscillators that pulse, guns that fire, and (astonishingly) structures that compute. This post is about the Game of Life: its rules, its menagerie of patterns, and why it became computer science's beloved emblem of emergence.

Four rules, a 2D grid of alive/dead cells, no players — and out come gliders that travel, oscillators that pulse, guns that fire, and structures that compute. Conway's Game of Life: its rules, its menagerie of emergent patterns, and why it became computer science's beloved emblem of emergence.

Grid Reliability and Assets

Balancing supply and demand keeps the grid running moment to moment — but the grid also runs on millions of physical assets (transformers, lines, substations) that age, wear, and occasionally fail, sometimes catastrophically. Keeping the grid reliable over time means anticipating and preventing those failures, not just reacting to them. This is a data problem — reams of sensor readings hinting at trouble before it happens — and it's where AI helps the grid stay reliable: predicting failures, spotting anomalies, and monitoring the vast physical system.

Balancing keeps the grid running moment to moment — but the grid also runs on millions of physical assets that age, wear, and occasionally fail catastrophically. Keeping the grid reliable means anticipating failures, not just reacting. That's a data problem, and it's where AI helps: predicting failures, spotting anomalies, and monitoring the vast physical system.

providers · Anthropic (Claude)

The same agent primitive as the Foundry lessons, backed by the Anthropic (Claude) provider instead of Azure AI Foundry.

A provider swap onto Anthropic Claude: build an anthropic.Client, pass it to anthropicprovider.NewAgent, and run the identical agent surface.

PostgreSQL Row-Level Security Is HIPAA Defense in Depth

PostgreSQL row-level security as HIPAA defence in depth. Why fail-open application filtering isn't enough, and how 'append-only at DB GRANTs' carries more of the §164.312(b) burden than people realise.

Elementary Cellular Automata — The Simplest Possible Rules

To understand cellular automata, start with the simplest possible version: a single line of cells, each either black or white, each updating based only on itself and its two immediate neighbors. These elementary cellular automata are about as minimal as a system can be — and yet, astonishingly, some of them produce patterns of unlimited complexity. This post is about the simplest CAs, how their 256 rules are numbered, and the shocking discovery that minimal rules can be maximally complex.

The simplest CA: a 1D line, two states, nearest-neighbor rules — minimal enough that there are exactly 256 rules, each named by a Wolfram number. The shock, found by enumerating all 256: behaviors range from boring to chaotic (Rule 30) to complex (Rule 110), and complexity appears even among the very simplest rules.

Demand-Side Flexibility

For a century, grid operation had one basic move: adjust supply to follow demand. The renewable era adds a second, transformative move — adjust demand to follow supply. If you can shift when electricity is used to when clean power is abundant, you turn demand from a fixed constraint into a flexible resource that helps balance the grid. Orchestrating that flexibility across millions of devices and distributed resources is a massive coordination problem, and it's one of the most exciting frontiers for AI in energy.

For a century, grid operation had one basic move: adjust supply to follow demand. The renewable era adds a transformative second move — adjust demand to follow supply. If you can shift when electricity is used to when clean power is abundant, demand becomes a flexible resource that helps balance the grid. Orchestrating that across millions of devices is a massive AI coordination problem.