developmentAI-assistedtransparency
One Developer, 250,000 Lines, and a Lot of Help
Johnathon Rhoades
AI agents wrote most of the code. Every line was governed. Here's what that means and why it matters.
Blog
data-sovereigntyvendor-lock-inphilosophy
Your AI's Memory Shouldn't Belong to a Vendor
Johnathon Rhoades
If you switched from ChatGPT to Claude tomorrow, what would you lose? Every custom instruction, every preference, every decision. That's vendor lock-in.
governancesecuritytransparency
The Day Our Firewall Broke (And What It Cost Us)
Johnathon Rhoades
The context gate had a bug that prevented agents from editing files. The agents couldn't fix it — because the firewall was working. Here's what happened.
governancearchitecturetesting
121 Agent Sessions, Zero Architectural Drift
Johnathon Rhoades
The codebase grew 9x over 121 AI agent sessions. Not one of them broke a founding architectural rule. Here's how.
origingamingphilosophy
What a Tabletop RPG Taught Me About AI Governance
Johnathon Rhoades
In April 2023, I tried to play Fantasy Age with ChatGPT. It couldn't follow the rules. That frustration became a two-year project.
announcementarchitecturegovernance
Introducing TVN Bridge: AI That Can't Disobey Your Rules
Johnathon Rhoades
We built a local-first governance platform that enforces constraints AI assistants cannot bypass. Here's why, and how it works.
AIGovernanceData PrivacyData SecurityTrustPhilosophy
Why We Govern Actions, Not Thoughts
Johnathon Rhoades
AI safety focuses on controlling what models think. TVN Bridge controls what they can do — deterministic enforcement at the action boundary, not the reasoning layer.
Showing 7 of 7 posts.