Research & Writing

Security and safety research, technical deep-dives, and notes from the lab.

KICR CCDC 26 Debrief: Deploying Autonomous Cyber Agents Under Real Competition Constraints

At KICR CCDC 26, ARIMLABS deployed autonomous AI cybersecurity agents in a live CCDC-style competition and finished 2nd overall. A technical debrief of what we deployed, what failed, and what surprised us.

ARIMLABS2026-02-25Competitions, Agents, Security

Read featured →

New2026-01-28

Copyright in AI Training Data: Why You Can Be Sued - and How to Avoid It

Copyright risk in AI training data can't be solved with policies alone. This article explains why enforceable, pre-training controls matter and how ARIMLABS approaches the problem.

New2026-01-26

AI Governance Is Failing, what can we do?

ARIMLABS conducted a cross-jurisdictional analysis of AI governance frameworks, revealing why document-centric governance breaks down in autonomous and agentic AI systems.

Closing the AI Governance Implementation Gap

ARIMLABS has created an extensive mapping of two major AI compliance frameworks - AICM and the NIST AI RMF - identifying how their controls align, overlap, and where gaps exist.

ARIMLABS R&D Team2025-11-16Governance

Read →

The Hidden Dangers of Browsing AI Agents

ARIMLABS uncovers key vulnerabilities in LLM-based browsing agents and proposes the first end-to-end threat model with practical defenses.

ARIMLABS R&D Team2025-05-19Agents, Security

Read →

Bypass of CVE-2023-44467: RCE in langchain PALChain

ARIMLABS successfully exploited a class pollution technique to bypass LangChain's security measures and achieve remote code execution.

ARIMLABS R&D Team2025-05-05Security

Read →

Research & WritingResearch & Writing

KICR CCDC 26 Debrief: Deploying Autonomous Cyber Agents Under Real Competition Constraints

Copyright in AI Training Data: Why You Can Be Sued - and How to Avoid It

AI Governance Is Failing, what can we do?

KICR CCDC 26 Debrief: Deploying Autonomous Cyber Agents Under Real Competition Constraints

Copyright in AI Training Data: Why You Can Be Sued - and How to Avoid It

AI Governance Is Failing, what can we do?

Closing the AI Governance Implementation Gap

The Hidden Dangers of Browsing AI Agents

Bypass of CVE-2023-44467: RCE in langchain PALChain

Closing the AI Governance Implementation Gap

The Hidden Dangers of Browsing AI Agents

Bypass of CVE-2023-44467: RCE in langchain PALChain

Research & Writing