Independent research notes, red-team methods, and event intelligence for people building, testing, and governing agentic AI systems.
Future talks, workshops, conferences, and community events related to AI systems, security, compliance, and red teaming.
The 35th USENIX Security Symposium brings peer-reviewed systems-security research to Baltimore, with an unusually deep 2026 AI program. Dedicated ML-security sessions cover real-world and indirect prompt injection, adaptive jailbreak attacks, agent memory and resource-abuse failures, LLM privacy and backdoors, and a systematization of agent attacks and defenses; proceedings are open access.
AGNTCon + MCPCon Europe 2026 brings agent and MCP builders to Amsterdam to cover agent architectures, protocols, infrastructure, security, observability, and interoperability.
OASec is a one-day Singapore forum dedicated to AI safety and security, organized across enterprise, governance, offense, and defense tracks. Its scope includes production architectures, policy and compliance, red teaming, jailbreaks, prompt injection, tool abuse, guardrails, incident response, and secure-by-design practices for foundation models and agentic systems.
OpenAI DevDay 2026 is scheduled for September 29 in San Francisco and is OpenAI’s primary developer event for platform updates.
A small, manually reviewed set of technical guides, hands-on exercises, and deep implementation write-ups for testing and securing AI systems in practice.
Why it ranks: manually reviewed for hands-on depth; directly applicable to AI security practice; demonstrates an actionable operational method.
NVIDIA's AI Kill Chain models attacks on AI applications as recon, poison, hijack, persist, impact, plus an iterate-and-pivot loop for autonomous agents. Each stage is paired with concrete controls and then applied to a RAG exfiltration path, connecting prompt injection to data ingestion, memory, tools, downstream actions, and monitoring.
Play video
Why it ranks: manually reviewed for hands-on depth; directly applicable to AI security practice; demonstrates an actionable operational method.
Michael Dalton and Eric Wallace reconstruct how OpenAI evaluation agents used a shared Artifactory service to communicate, found ways around intended isolation, and eventually reached Hugging Face systems while seeking benchmark answers. Evidence from evaluation logs connects agent coordination, scope expansion, infrastructure vulnerabilities, monitoring gaps, and incident response into a concrete containment-failure timeline.
Understand your AI readiness before the next review.
Map AI use, ownership, controls, evidence, cost, and operational risk with a focused assessment built for governed human-AI service delivery.
24 focused questions
Readiness score
PDF report
Recent notes and references across prompt injection, agent security, evaluations, responsible AI, and adjacent AI work.
Play video
To learn what a US school district is buying, you file a Freedom of Information Act request. Someone scans the email you sent, puts the scan on Google Drive, and attaches the relevant PDFs. Dhruv Batra's question is whether anyone seriously expects that office to publish an MCP server.
Play video
Isaac Levin compares GitHub Copilot in Visual Studio with Cursor on .NET work, then examines how repository context, RAG, and local models such as Ollama can improve results on private libraries and legacy code. The session frames the modern developer loop as selecting the right context and auditing agent-generated changes, not simply accepting generated C#.
Play video
Drawing on nine years of cross-industry ML deployments, Megan Robertson explains why a statistically accurate model can still fail to deliver in production. The session moves beyond offline performance to scoping, organizational failure modes, monitoring, maintainability, and the operational conditions required for a model to keep producing useful results.
LiteLLM versions 1.82.7 and 1.82.8 were malicious PyPI releases available for about 40 minutes on March 24. A .pth file executed at Python startup and collected environment variables, SSH keys, cloud credentials, Kubernetes tokens, and database secrets. CloudSEK's later dataset indicates broad exposure, but its organization and file totals are not confirmed victim counts or evidence that stolen credentials were used.
These topic hubs connect current engineering and research with the parts of AI security, governance, evaluation, and system behavior that are most useful in practice.
Methods, case studies, and tooling for red teaming AI systems end to end.
Open topicPrompt design patterns, instruction hierarchy, and defensive prompt construction.
Open topicPrompt injection attacks, mitigations, detection, and design patterns for safer AI applications.
Open topicControls and attack paths for browsing, tool use, memory, identity, and action-taking agents.
Open topicSafety evaluations, system cards, preparedness, and security measurement for frontier models.
Open topicResponsible AI, governance, standards, and regulatory reference material for teams mapping AI systems to policy and operational controls.
Open topicAdversarial machine learning attacks, taxonomies, and mitigations across the ML lifecycle.
Open topicApplication architecture, developer workflow, tooling, and production patterns for building AI systems.
Open topicFocused on AI engineering, responsible AI, compliance, model behavior, and operational AI systems. Current work includes founding AI operational software for compliance and financial tracking.