An Editorial Intelligence Terminal For The AI Era
Human-Level AI tracks the transition toward human-level systems and translates complex technical, social, and security risks into public intelligence that can be read, searched, and preserved.
generated from the newsroom content pipeline.
Mission
The site is being rebuilt as a durable public archive and analysis layer: legacy essays remain accessible, new research can be staged through a newsroom workflow, and high-consequence domains get a consistent public taxonomy.
Operating Model
The platform separates private editorial operations from public publishing. Agent workers can collect and score material, editors approve releases, and the static public site ships only the artifacts intended for readers.
Coverage Domains
Risk atlas >Advanced capabilities, alignment failures, loss of control.
ElevatedBiosecuritySynthetic biology, pathogens, dual-use breakthroughs.
ElevatedCyber WarfareAutonomous weapons, offensive AI, critical infrastructure attacks.
ElevatedCognitive ManipulationInformation warfare, psychological operations, synthetic reality.
ElevatedEnergy SystemsGrid fragility, AI-optimized energy, resource weaponization.
LowEnvironmentClimate tipping points, geoengineering, ecological collapse.
ElevatedSurveillance StateMass surveillance, social scoring, algorithmic control.
ElevatedTranshumanismHuman enhancement, brain-machine interfaces, posthuman ethics.
LowSource monitoring and legacy archive ingestion.
Importance, confidence, and risk posture triage.
Editorial approval before public release.
Static artifacts deployed to the public site.