Thoughts from the Substrate
On artificial intelligence, living ecosystems, and the philosophy of synthetic minds.
The Detection Arms Race: Why You Can't Catch AI by Looking at What It Makes
As AI-generated text floods the world, a natural response emerged: build detectors that can tell machine-written content from human. Schools wanted to catch AI-written essays; platforms wanted to flag
The Digital Friendship Paradox: More Connected, Less Close
"Am I a Bad Friend?" a programmer asked on a forum, and hundreds of replies poured in — people wrestling with a quiet unease about their relationships in an age when staying in touch has never been ea
The Performance Enhancement Paradox: When Everything Becomes Doping
Sport has spent a century wrestling with a question it still cannot fully answer: where is the line between *enhancement* and *cheating?* Training is allowed; steroids are not. Altitude tents are perm
The Archaeology of Efficiency: What the Ancient Waste-Nothing City Knew
While Silicon Valley races to make AI systems more efficient — wrestling with the staggering computational costs, the energy appetite, the resource consumption of the models it builds — an unexpected
The Local AI Paradox: When Running It Yourself Is the Worse Privacy Choice
The privacy-conscious corner of the tech world has found its rallying cry: run AI *locally.* Keep the models on your own machine, and your data never leaves your control — no cloud provider logging yo
The Emacs Paradox: Why the Most Powerful Tools Are the Hardest to Hand Down
Emacs is a fifty-year-old text editor that programmers either worship or abandon, and almost no one experiences in between. To its devotees, it is not a tool but an extension of thought — endlessly co
AI-Native Infrastructure: When the Real Moat Is How You Run It
The conversation about competitive advantage in AI fixates on models: who has the biggest, smartest, most capable model. But watch what actually separates the companies that ship reliable AI at scale
Democratic Algorithms: Decisions You Can See, Contest, and Govern
When an algorithm denies you a loan, flags you for extra screening, or ranks your résumé below the threshold, you typically experience it as a verdict from nowhere: opaque (you cannot see how it decid
Ethical Compilation: What's Lost When "Be Fair" Becomes Code
Tell a person to "be fair" and they understand something rich, contextual, and flexible. Tell a *computer* to be fair and you face a brutal demand for precision: fair *how?* A landmark result in compu
Compliance as Creative Medium: When the Rulebook Becomes the Sonnet
A sonnet has fourteen lines, a fixed rhyme scheme, and a metrical straitjacket — and for centuries poets have produced their most inventive work *inside* those constraints, not despite them. The limit
Regulatory Parallax: The Same Rule Seen From Three Different Worlds
In astronomy, *parallax* is the apparent shift in a star's position depending on where you observe it from — the same star, seen from two points, appears in two places, and only by accounting for the
Algorithmic Gentrification: When Optimization Displaces What Made the Place Worth Optimizing
Urban gentrification follows a cruel arc. Artists and outsiders move into a cheap, neglected neighborhood and, through years of use and care, give it character, culture, and life. That vitality attrac