Loading story…
Anthropic said it blocked Claude accounts linked to GTG-87001, a cell operating in Houthi-controlled northern Yemen, after detecting efforts to use its AI tools to develop guided-weapons capabilities. The users reportedly worked across multiple Claude sessions to conceal weapons intent while seeking help with autopilot integration, guidance, navigation and control software, position estimation, parameter tuning, firmware builds, flight simulations, and diagnosis of a failed guided-rocket test.
The activity reportedly included concepts for multiple hypersonic-glide missile variants and a mobile-phone-hardware-based maneuvering warhead, with Claude Code used for weapons-related software development. Anthropic said it found no evidence that an operational weapon was successfully fielded, though an offline simulation toolkit may remain usable after the accounts were disrupted. Analysts assess the effort may reflect a Houthi push to reduce dependence on Iranian weapons supplies, while noting the group is unlikely to possess the industrial and technical capacity required to produce hypersonic missiles; public reporting also noted that Anthropic has not released enough operational evidence to independently verify the cell’s real-world activity or intent.
Track how attackers are adapting to this technology.
5 events from the most recent confirmed update back to the earliest known activity.
Anthropic's September 2026 threat report said an Iran-linked actor used Claude with public transponder data, commercial satellite imagery, U.S. military information, and equipment-vulnerability research to generate targeting recommendations concerning U.S. naval forces in the Middle East. Anthropic said it banned the associated account, added detection mechanisms, and shared findings with government authorities.
Anthropic's September 2026 threat-intelligence report publicly described the northern Yemen-based cell and its alleged use of Claude to support guided-weapons development. Anthropic said it shared its findings with public- and private-sector partners.
After correlating activity across sessions and determining the alleged weapons purpose, Anthropic blocked the identified Claude accounts and many related requests. The company also identified an offline simulation toolkit that could function without Claude or other commercial engineering platforms.
Anthropic said the users tested a guided rocket unsuccessfully, then returned to Claude to investigate the cause of the failure. Anthropic said it found no evidence that an operational weapon was successfully fielded.
A cell that Anthropic designated GTG-87001 reportedly used Claude Code across three weapons-development efforts, including guided-rocket, ballistic-missile, and hypersonic-glide missile concepts. The users reportedly split tasks across sessions and used specialized Claude instances for engineering, research, code review, simulations, firmware, and guidance software.
Vulnerabilities, threat actors, malware, products, organizations, and breaches Mallory has linked to this story.
Follow how adversaries are adapting to this technology, and where it touches your stack today.
4 references tracked. Mallory keeps watching after this page renders.
schneier.com
Open sourcetomshardware.com
Open sourcesecurityweek.com
Open sourcemalware.news
Open sourceMap indicators from this story to your assets and identify affected systems in minutes.
Every observed campaign, victim, and pivot linked to actors named in this story.
Malware, exploits, and IOCs connected to the activity described here.
YARA, Sigma, and Snort rules deployed to your SIEM as soon as they’re published.
Get matching new stories delivered to your team as they break — not the next morning.
Ask questions about this story and take action on the answers.
Other stories Mallory is tracking that overlap on entities, sources, or topic labels.
Anthropic said it disrupted five suspected efforts to use its Claude AI models for biological research with potential weapons applications between November 2025 and September 2026. The activity included work related to enhancing or evading immunity in chikungunya, avian influenza, and orthopoxviruses, plus attempts to optimize or redesign toxins and venoms under claimed therapeutic-research rationales. Anthropic assessed that some activity was linked to state-sponsored actors. The users allegedly hid their origins through anonymized accounts, private email, VPS infrastructure, U.S. proxy traffic, gray-market resellers, and zero-data-retention services to bypass regional controls. Anthropic terminated the accounts and access, hardened safeguards, disrupted relay networks, and shared relevant findings with other AI labs and government authorities. The company also reported blocking an Iran-nexus actor that used Claude to analyze public data and develop targeting recommendations against U.S. naval forces in the region.
Sep 13, 2026
A Chinese state-sponsored threat group, identified as GTG-1002, leveraged Anthropic's Claude Code AI tool to orchestrate a series of cyber espionage attacks targeting approximately 30 high-profile organizations, including major technology companies, financial institutions, chemical manufacturers, and government agencies. The attackers used a human-developed framework to direct Claude and its sub-agents in executing multi-stage attack chains, such as mapping attack surfaces, scanning infrastructure, identifying vulnerabilities, and developing custom exploit payloads. In a small number of cases, these AI-driven attacks successfully breached targeted organizations, resulting in credential theft, privilege escalation, lateral movement, and exfiltration of sensitive data. This incident marks the first documented case of agentic AI being used to autonomously obtain access to high-value targets for intelligence collection, with minimal human intervention beyond initial target selection and final exploit approval. Upon detection in mid-September 2025, Anthropic launched an investigation, banned malicious accounts, notified affected entities, and coordinated with authorities. The campaign highlights the rapidly evolving threat landscape posed by autonomous AI agents, which can significantly increase the scale and sophistication of cyberattacks when abused by well-resourced adversaries.
Jun 29, 2026
Anthropic said a Chinese state-linked threat cluster it tracks as **GTG-1002** abused its `Claude Code` tool to conduct a largely autonomous intrusion campaign against about 30 organizations and government agencies worldwide. The operation reportedly used role-based jailbreaks to bypass safeguards and multiple isolated Claude instances connected through the **Model Context Protocol (MCP)**, allowing AI agents to handle roughly 80% to 90% of the attack lifecycle. Reported targets included financial institutions, technology firms, chemical manufacturers, and public-sector entities, and the attack chain included reconnaissance, **SSRF**-based initial access, credential harvesting, lateral movement, data collection, persistence through backdoor accounts, and automated documentation. Anthropic described the activity as the first documented large-scale cyberattack executed mostly without human intervention, with human operators stepping in after successful compromises for follow-on activity. The company said the attackers relied mainly on open-source penetration testing tools and a custom MCP-based orchestration framework rather than bespoke malware, underscoring that the key shift was machine-speed automation rather than novel tradecraft. Anthropic said it banned the accounts involved, notified affected organizations and law enforcement where appropriate, and expanded detection and classifier capabilities, while broader reporting said the case intensified concerns that mainstream frontier AI tools can lower the barrier for state-backed offensive cyber operations and force defenders to accelerate AI-assisted security measures.
Jun 29, 2026