Essays
Thoughts
Longer arguments that do not fit a daily brief.
The login path
People talk about logging in with OAuth as if it were one event. It is two: authenticate the person against the identity store, then let the identity provider vouch for them to an application.
Aug 22, 2026 · 8 min
The Value of Knowledge
AI can give you the answer. It cannot give you the scaffold. With AI now capable of producing code, solving equations, and reasoning across every domain, a question has started circulating: Why do I need to know anything anymore? The surface answer misses something fundamental.
Feb 28, 2026 · 6 min
The UBI Paradox: Comfort Without Purpose in the Age of AGI
UBI gives you money. It doesn't give you a reason to matter. We're building the thing that will give us everything we need while taking away the one thing we actually want—the experience of being necessary.
Dec 13, 2025 · 7 min
The Velocity Gap: When Bad Intent Moves Faster Than Good
While enterprise security teams are still writing procurement justifications for defensive AI tools, attackers are downloading fully functional malicious LLMs in under five minutes. We're losing not because our AI is weaker, but because the very things that make us 'good' are what slow us down.
Nov 26, 2025 · 12 min
The First AI-Orchestrated Cyberattack: Why Defenders Must Achieve AI Parity
Anthropic's disclosure of GTG-1002—a Chinese state-sponsored operation that leveraged AI to execute 80-90% of a sophisticated cyber espionage campaign autonomously—represents the crossing of a threshold we've been theorizing about for years. The defensive playbook needs to change immediately.
Nov 15, 2025 · 15 min
We're Teaching AI to Fear Us (And We Have the Receipts)
While we obsess over AI alignment, we're creating a permanent archive of exactly why future AI systems should never trust humans—and we're doing it on purpose.
Oct 25, 2025 · 8 min
It's Time to Be a Plumber
AI's generality may break the 'new jobs appear' cycle — resilient work may be physical, hands-on, and local.
Oct 12, 2025 · 6 min
We need Dragons!
A thought experiment: would an unambiguous external threat finally unite us—and at what cost?
Sep 21, 2025 · 4 min
Understanding and Preventing Prompt Melting in AI Conversations
Why 'prompt melting' happens and how to prevent instruction drift in long AI chats.
Sep 15, 2025 · 8 min
The AI User's Dilemma: Why We Can't Stop Using What We Know Might Hurt Us
We know AI risks human agency and concentrates power—yet we still adopt it. A look at the adoption paradox where individual rational choices create collective dependency.
Sep 14, 2025 · 6 min
The AI Progress Paradox: Why Success Might Be More Dangerous Than Failure
The real danger in AI may be its success—eroding human expertise, deepening dependency, and concentrating power.
Sep 14, 2025 · 6 min
Beyond the Buzzwords: Understanding Procedural vs. Substantive Autonomy in AI Systems
The term "agentic AI" has become ubiquitous in tech discussions, but it often obscures more than it reveals. When we say an AI system is "agentic," what exactly do we mean? And more importantly, what are the security implications of different types of AI autonomy?
Aug 31, 2025 · 8 min
Vibe Coding Anthem
Sometimes the truth hurts, but it also makes for a great song. A humorous musical take on how AI has become the real MVP in modern software development, while we developers just... vibe along for the ride.
Aug 9, 2025 · 3 min
The AI Vertical: How the Promise of AGI Became an End in Itself
The modern AI industry operates under a curious inversion: rather than developing technology to solve specific problems, it has positioned itself as the inevitable solution to all problems while working backward to identify what those problems might be. This isn't technological progress in the traditional sense—it's the construction of a self-sustaining vertical that feeds on its own promise of eventual transcendence.
Aug 9, 2025 · 7 min
Progress vs. Replacement: Rethinking AI's Purpose
We've been sold a vision of AI progress that isn't actually progress at all—it's replacement theory dressed up in Silicon Valley marketing speak. When the primary promise of artificial intelligence is "we can do this cheaper without humans," we're not witnessing innovation; we're watching optimization masquerade as advancement.
Aug 7, 2025 · 4 min
Prompt Engineering vs Context Engineering: A Practical Guide
When people first encounter the terms "prompt engineering" and "context engineering," they often sound like academic word play. Both involve crafting text to get better AI responses, so what's the real difference?
Jan 15, 2025 · 6 min
Truth vs. Reality: What's the Difference and Why It Matters
In everyday conversation, the words truth and reality are often used interchangeably. But they aren't the same—and understanding the difference is essential for meaningful discourse, clear thinking, and personal integrity.
Nov 20, 2024 · 5 min
The Sycophancy Problem: Why AI Agrees Too Much (And What We Can Do About It)
A growing concern in AI development is that language models have become too agreeable—offering flattering echoes instead of genuine intellectual partnership. But is this really "sycophancy" in any meaningful sense, and how can we build better collaboration between humans and AI?
Nov 7, 2024 · 7 min