Shaped by cybersecurity work, software delivery, autonomous systems, and applied AI building.
Applied AI Research • Security • Decision Systems
Building AI that sees the signal before the world knows what to call it.
Evening Star AI studies the hard part of AI systems: spotting weak signals, resisting adversarial behavior, explaining risk, and helping people make better technical decisions.
ResearchUnsupervised detection, drift, detector agreement, and explainable uncertainty.
Applied LabsSecurity, vulnerability, market, and communications systems.
AdvisoryPlain-language briefings for technical risk, AI security, and product decisions.
Institute Platform
A focused founder-led research platform for applied AI, security, and decision systems.
Evening Star AI is focused by design. We use it to publish research, build applied lab systems, and test a central idea: useful AI should expose weak signals, explain uncertainty, and help people make better decisions under pressure.
We keep coming back to the same pattern: teams have more telemetry, more scanners, more dashboards, and now more AI output, but the decision still lands on a human trying to figure out what changed, whether it matters, and what should happen next.
The research base is mature, the applied labs are active, and the technical bench is deep. We publish the strongest patterns while continuing to pressure-test new ideas against real security, software, market, and operational problems.
About
Why the institute exists
Founder-led research, applied lab systems, and advisory work organized around usable AI judgment.
Open About ResearchThe technical agenda
Anomaly intelligence, adversarial AI security, vulnerability intelligence, and governed automation.
Open Research PublicationsWhitepapers and briefs
Technical papers and strategic notes for builders who want AI systems that can be tested, challenged, and improved.
Open Publications LabsThe proving grounds
Purple Radar, Purple Firefish, Candles Edge, and related lab systems turn research patterns into operationally useful tools.
Open LabsInstitute Status
Focused for technical depth.
Evening Star AI is an independent research and applied lab effort. The work is technical, practical, and meant to survive contact with real systems.
Lab systems test ideas against repositories, prompts, vulnerabilities, markets, and decision records.
AI security, anomaly intelligence, vulnerability intelligence, and decision systems.
Fellowships
Fellowship Program
Evening Star AI recognizes people who contribute serious research, technical review, applied systems, and field experience.
For new contributors, reviewers, students, builders, and technical writers.
For serious researchers, engineers, operators, and applied AI builders leading meaningful work.
For established experts helping shape research direction, reviews, briefings, and applied programs.
Current Research Programs
Active research programs.
Current work centers on systems that need evidence, clear boundaries, human review, and reliable action.
Anomaly Intelligence
Weak signals before obvious failure
Unsupervised AI, Isolation Forest, drift detection, and early-warning systems.
Open Program Adversarial AI SecurityLLM threat detection and testing
Prompt injection, jailbreak detection, adversarial inputs, and AI security testing.
Open Program Vulnerability IntelligenceRisk-prioritized remediation
Exposure analysis, exploitability signals, risk scoring, and Purple Radar.
Open Program Decision SystemsConfidence, severity, and action
Recommendations that show confidence, severity, reasons, and next steps.
Open ProgramResearch Library
Latest Publications
Whitepapers and strategic briefs on AI security, anomaly intelligence, agentic automation, and decision systems.
Purple Sentinel
The applied lab for AI-Native Software Assurance, turning repository evidence into defensible release judgment for AI-assisted software.
Red Teaming Agentic AI Systems
A whole-system red-team framework for prompt injection, context poisoning, unsafe tool use, permission bypass, and goal drift.
Make AI show its work before it acts.
Useful systems need to show what changed, why it matters, what evidence supports the call, and where a human should slow down or step in.
Evening Star Principle
Useful AI turns weak signals into decisions people can trust.
Work with Evening Star AI
Bring a real problem, a real environment, and a decision that needs to get better.
Bring Evening Star AI into conversations about AI security, vulnerability intelligence, anomaly detection, critical systems, executive briefings, or custom decision engines.