Topic
MITRE ATLAS
IT Elements 2026 to Stage IT Apocalypse Simulations and Critique Russian Vendors
Ransomware & Extortion
Aligning AI Adoption Maturity with AI Security Using CMMI and Russian Regulatory Requirements
AI Security
Adam Shostack Presents PHANTOM-B Threat Modeling Framework for LLMs at Black Hat USA
AI SecurityAI Agents Given Code and API Access Can Now Assist Attackers
An AI assistant that only answers questions can make mistakes, but an AI agent with access to email, code execution, corporate APIs and internal data can make those mistakes inside production infrastructure. The difference is fundamental: once tools, credentials and internal data are connected to the model, it becomes a privileged user that may not distinguish legitimate commands from hidden instructions on a web page. OWASP lists prompt injection, sensitive data disclosure, unsafe output handling and excessive autonomy as key risks for LLM applications. MITRE ATLAS specifically describes techniques involving prompt injection, context poisoning and tool invocation by AI agents. The article examines how agents differ from chatbots, how attackers can control them through untrusted content, and why a system prompt alone cannot protect code, data and APIs. CyberED is running its free NeuroAugust series of events and materials on AI in cybersecurity, including a session on secure AI system development.
Bank of Russia Publishes Methodological Recommendations No. 3-MR on AI Security for Financial Market Participants
The Bank of Russia has released methodological recommendations No. 3-MR dated 16 June 2026, providing detailed guidance on ensuring information security during the development and use of artificial intelligence systems in the financial sector. The document builds on the earlier Code of Ethics for AI in finance and integrates with existing risk management, operational resilience, and data protection frameworks already familiar to credit institutions and other market participants. It introduces standardized terminology for AI-specific threats such as hallucinations, data drift, and poisoned datasets while outlining six risk categories and a four-stage AI system lifecycle model. Organizations are advised to apply threat modeling based on FSTEC methodology, implement proportional controls across data preparation, development, training, and operation phases, and maintain human oversight for high-risk automated processes. Special attention is given to supply chain risks involving third-party vendors and open-source components, requiring due diligence, provenance tracking, and contractual safeguards aligned with existing outsourcing standards. The recommendations remain non-binding yet signal clear regulatory expectations that are likely to influence future compliance checks and audits.