Anthropic
Detecting and Countering Misuse of AI (2026) - Documents adversary attempts to exploit frontier foundation models for automated cyber operations and state-sponsored espionage. Threat intelligence observations indicate that threat actors utilized automated prompt workflows to…
- from
- Annual Security Reports
- added
- 2026-10-19
- likes
- 0
similar
-
Anthropic anthropic.com
Threat Intelligence Report (2025) - Examines how threat actors are leveraging advanced AI models, particularly Claude, to escalate and refine their cyber operations. Key findings demonstrate AI is weaponized as an active agent in sophisticated attacks like vibe hacking for data…
-
Anthropic menlovc.com
The State of AI (2026) - Examines the current landscape of AI agent adoption and deployment across enterprises, based on a survey of over 500 technical leaders. Key findings reveal that 57% of organizations deploy agents.
-
Anthropic anthropic.com
Provides access to a class of advanced large language models called Claude designed to provide human-like text responses and engage in conversational interactions.
-
Anthropic Prompt Engineering docs.anthropic.com
Techniques for defining success criteria, testing prompts, and improving model behaviour.
-
Google cloud.google.com
Advances In Threat Actor Usage Of AI Tools (2025) - Focuses on the increasing use of AI tools by threat actors, highlighting a shift towards deploying novel AI-enabled malware in active operations. Key findings include the emergence of malware families using LLMs for dynamic…
-
Anthropic resources.anthropic.com
Agentic Coding Trends Report (2026) - Examines the evolution of agentic coding in 2026, focusing on the transition from individual AI assistants to coordinated multi-agent systems that manage complex software development lifecycles. Research indicates that while developers…
Annual Security Reports › Analysis Reports > AI and Emerging Technologies: “Detecting and Countering Misuse of AI (2026) - Documents adversary attempts to exploit frontier foundation models for automated cyber operations and state-sponsored espionage. Threat intelligence observations indicate that threat actors utilized automated prompt workflows to…”