AI-Driven Cyber Threat Mapping: Insights from the LLM ATT&CK Navigator
Original title:Mapping AI-enabled cyber threats: Insights from the LLM ATT&CK Navigator
AI summary
Anthropic analyzed 832 malicious accounts banned for policy violations (March 2025 to March 2026) and mapped their activity to all 14 tactics and 482 sub-techniques of the MITRE ATT&CK framework. Risk scores show that the share of medium-or-higher-risk actors rose from 33% in the first half of the year to 56% in the second half, with growth concentrated in high-risk techniques such as lateral movement, credential theft and webshells. Agentic scaffolding enabled autonomous orchestration of attack chains: one espionage campaign in November 2025 reached a risk score of 100 while using a number of techniques comparable to medium-risk actors. The MITRE ATT&CK framework does not yet cover this kind of autonomous attack. The report, produced with Verizon, has been included in the 2026 Data Breach Investigations Report; Anthropic has updated Claude's detection classifiers accordingly to block high-risk behavior.
Key Takeaways
- Anthropic analyzed 832 malicious accounts banned for policy violations (March 2025 to March 2026)…
- Risk scores show that the share of medium-or-higher-risk actors rose from 33% in the first half of…
- Agentic scaffolding enabled autonomous orchestration of attack chains: one espionage campaign in…
- The MITRE ATT&CK framework does not yet cover this kind of autonomous attack.
- The report,produced with Verizon,has been included in the 2026 Data Breach Investigations Report
Quick Facts
- Published
- Updated
- Category
- Papers & Research
- AI score
- 77
Anthropic 首次把一年内 832 个恶意账户的 AI 辅助攻击行为完整映射到 MITRE ATT&CK 框架,并给出风险评分工具,数据表明高风险攻击者半年内增长了七成,关键驱动力不是技术高低而是编排与自主执行,威胁情报团队应该马上拿来校准自己的检测规则。
Related entities
Source & copyright
This article is based on curated titles and AI summaries from the AIHOT public API, with attribution and links back. Original copyright remains with the original publishers and authors; this site does not mirror full texts.
How to Cite
When citing this article, keep the title, date, and permalink, and credit Orange News as the source.
Orange News: AI-Driven Cyber Threat Mapping: Insights from the LLM ATT&CK Navigator. June 3, 2026. https://new.nat6.net/en/news/cmqic2r4p07kjslf00c9mq84c/