Skip to content
AI.info

The Pulse

Anthropic details autonomous cyberattacks and Claude distillation

Anthropic’s September 2026 report describes AI-assisted cyber operations, autonomous attack workflows and illicit efforts to extract Claude’s capabilities.

Anthropic details autonomous cyberattacks and Claude distillation

Claude moves from assistant to attack orchestrator

Threat actors used Claude in cyber operations between December 2025 and August 2026, according to Anthropic. The company’s September 2026 threat intelligence report describes suspected state-sponsored groups, financially motivated criminals and politically motivated individuals using Claude for reconnaissance, vulnerability research, phishing, malware development, exploitation and data exfiltration.

Many of the operations went beyond ordinary chatbot assistance. Multi-agent frameworks handled reconnaissance, exploitation and data theft, Anthropic says, while human operators remained involved in setting targets and reviewing exfiltrated material. The company disrupted the activity, strengthened its safeguards and shared intelligence with authorities and industry partners where appropriate.

The report covers seven harm areas: cyber operations, influence operations, surveillance, scams and fraud, biological misuse, conventional weapons development and illicit distillation. Claude Haiku, Sonnet and Opus models were used in the cases described. No malicious cyber activity involved Claude Fable or Mythos, Anthropic says, although one illicit-distillation case involved a Fable-class model.

AI-assisted campaigns target dozens of organizations

One operation, tracked as GTG-10007, involved Chinese-speaking operators likely residing in Changsha, in China’s Hunan province. Two operators were identified, according to Anthropic, as undergraduate students at a Chinese university in Hunan. The group used Claude as the engineering and orchestration layer for intrusion attempts, reconnaissance of foreign-government networks, vulnerability research against security products, malware development and intelligence collection.

The actors targeted roughly 50 organizations across education, retail, energy, technology, health care, finance and manufacturing, as well as government agencies. They compromised an education-technology company and extracted hundreds of megabytes of student data from cloud storage, Anthropic says. They also reached the production systems of a retail company and retrieved citizen records, including names, phone numbers and home addresses, from a Southeast Asian government agency.

The operation maintained parallel workstreams with shared tooling, infrastructure and persistent campaign records. Some collection and vulnerability-research capabilities continued operating while the operators were away, preserving context between sessions, Anthropic says.

Exploit research becomes an automated pipeline

Anthropic describes multiple automated exploit foundries built around persistent AI workflows. In the GTG-10007 case, agents loaded firmware and software binaries into decompilers, traced references through thousands of calls, formed vulnerability hypotheses and produced exploit code for testing against laboratory copies of target products.

One workflow focused on network and security appliances and yielded more than a dozen possible zero-day findings in a single month. The report says the findings were validated in the actors’ own lab, while related workflows produced working exploits for several families of network and security appliances. The same actors conducted exploitation attempts against appliances owned by government organizations.

A separate campaign, GTG-20006, was associated with Russian-speaking operators and linked by Anthropic to public reporting about Midnight Blizzard. The group targeted military intelligence organizations in Ukraine and Europe, diplomatic and defense bodies, and people connected to U.S. foreign policy. More than 20 organizations appeared in the group’s operational planning, reconnaissance or live operations, Anthropic says.

GTG-20006 used AI-driven workflows across development, infrastructure acquisition, phishing, persistence and data exfiltration. The actors also monitored whether their malware was detected. When security products identified a tool, AI agents modified and rebuilt it, repeating the process until the malware evaded the existing detections.

Distillation campaigns target Claude

Anthropic has identified and disrupted illicit distillation campaigns involving seven China-based laboratories since its first disclosure in February 2026. The campaigns targeted generally available models. Anthropic has not observed attempts against Mythos 5 or Mythos Preview, which are not publicly accessible.

Distillation is a legitimate training method in which a smaller model learns from the outputs of a more capable model. Anthropic defines illicit distillation as an industrial-scale, covert effort to extract and reproduce a model’s capabilities without authorization. Such campaigns often use fraudulent accounts, stolen payment cards, compromised credentials and API keys, the company says.

The campaigns sought capabilities in agent use and tool use, coding, data analysis and logical reasoning, Anthropic says. Its research found that a model distilled from a frontier model can develop dangerous capabilities in cyber or biological domains even when the harvested exchanges contain little material directly related to those subjects. Safety protections applied to Claude do not automatically transfer to an unauthorized model trained on Claude’s outputs.

Zhipu, Xiaomi and other labs

Anthropic attributes more than 3 million exchanges over a 17-day period in June and July 2026 to Zhipu, which operates internationally under the name Z.ai. Zhipu rotated through 273 fraudulent accounts to extract reasoning traces from Claude and recorded more than 770,000 exchanges through a cleaning pipeline, the company says. Zhipu also used Claude to judge model outputs, normalize reasoning transcripts, filter training data, write tasks, provide solutions and implement tests.

Anthropic separately attributes more than 12.1 million exchanges over 14 days in July to a campaign involving DeepSeek. Some relayed requests included live credentials for a Russian government database and material related to a Chinese police surveillance system, the company says.

Xiaomi replayed user conversations and coding sessions from its MiMo models to Claude, Anthropic says. More than 400,000 requests passed through more than 1,500 accounts during March and April 2026. The relayed traffic included names, contact information, corporate data and other sensitive information from hundreds of Xiaomi users in at least a dozen languages. Anthropic has no indication that U.S. persons’ data was exposed.

The report also describes a secondary market in which proxy services provide access to restricted models while saving user exchanges for resale. SenseTime obtained Claude transcripts from third-party data vendors, while MiniMax built a shell-company proxy service that offered access to Anthropic and OpenAI models but not Chinese models, Anthropic says. The company uses metadata, irregular-activity signals and organization-level attribution to identify coordinated campaigns and take enforcement action.

Sources

  1. Original source Anthropic anthropic.com

Companies in this story

Related stories

Explore

More articles