Claude in cyberattacks: how AI is changing hackers' work

Claude in cyberattacks: how AI is changing hackers' work

On September 10, 2026, Anthropic disclosed one of the most illustrative examples of generative AI being used in a real cyber operation. The GTG-20006 group, which the company assesses as linked to Russian state cyber espionage, used Claude not merely for writing code or consultations. AI was integrated almost…

Agent harness: how an AI agent's runtime environment works

Agent harness diagram with language model, tools, memory, and control

Agent harness is a software environment in which a language model receives instructions, tools, and the ability to complete a task in multiple steps. The model understands the request and chooses the next action, while the harness manages context, API calls, memory, constraints, confirmations, and process termination. Specifically…

AI agent memory: three layers, architecture, and risks

AI agent memory: three layers, architecture and risks

AI agent memory is not a hidden capability of the language model, but an external system for storing, retrieving, and updating context. It allows the agent to continue working between sessions, take user preferences into account, use accumulated facts, and avoid repeating the same mistakes. A practical architecture…