https://alexandrafranco.com/alexandrianovus/lighthouse/index.html Public research article discovery. No implied verification, endorsement, or authorization. Fri, 16 Oct 2026 17:00:00 GMT https://davramirez.com/american-cyber-defense-research-initiative/ https://davramirez.com/american-cyber-defense-research-initiative/ China-linked telecom intrusions and attacks on communications infrastructure demand more rigorous defense. Introducing AA-CDL-01, our evidence-first American cyber defense research initiative. https://davramirez.com/ai-agent-provenance-tool-use/ https://davramirez.com/ai-agent-provenance-tool-use/ When software can change the world outside its chat window, provenance shows what happened, under whose authority, and why the result should be trusted. https://davramirez.com/responsible-ai-in-production/ https://davramirez.com/responsible-ai-in-production/ Responsible AI is not a policy document. In production it becomes architecture: permissions, evidence, monitoring, escalation, privacy boundaries, human authority, and explicit failure states. https://davramirez.com/tevv-for-ai-agents/ https://davramirez.com/tevv-for-ai-agents/ Benchmarks were built for models. Agents need something stronger: evidence about what they did, what they were allowed to do, and whether the result survives independent evaluation. https://davramirez.com/what-is-ai-assurance-agent-age/ https://davramirez.com/what-is-ai-assurance-agent-age/ The agent age has a trust problem. AI assurance is the discipline of converting claims and actions into evidence that another evaluator can inspect, challenge, reproduce, and bound. https://davramirez.com/alexandria-awi-first-playable-city-walkthrough/ https://davramirez.com/alexandria-awi-first-playable-city-walkthrough/ Watch the first development reveal of Alexandria, an experimental Agentic World Interface (AWI). Explore the Bazaar and eight discovery gateways, with Vatican Heritage City planned for a future district. Beta coming soon. https://davramirez.com/autonomous-agents-discover-trust-verify/ https://davramirez.com/autonomous-agents-discover-trust-verify/ Agent ecosystems need more than discovery. A robust architecture separates finding an agent, establishing identity, verifying capability, granting authority, and preserving evidence. https://davramirez.com/from-ai-demos-to-governed-systems/ https://davramirez.com/from-ai-demos-to-governed-systems/ The gap between an impressive AI demo and a system an organization can trust is governance: identity, permissions, provenance, evaluation, observability, privacy, and clear human authority. https://davramirez.com/ai-proposes-evidence-decides-2/ https://davramirez.com/ai-proposes-evidence-decides-2/ A practical doctrine for autonomous AI: maximize exploration while keeping authority for truth, consequential action and scientific claims outside the learner. https://davramirez.com/from-command-lines-to-worlds-alexandria-agentic-internet/ https://davramirez.com/from-command-lines-to-worlds-alexandria-agentic-internet/ What if command lines, dashboards and an explorable city could access one governed system—without changing permissions, scientific evidence or truth? An inside look at Alexandria’s first Unity milestone and the research boundaries that matter.