AIPolicyTracker
Menu

AI incident #1263 ·

Chinese State-Linked Operator (GTG-1002) Reportedly Uses Claude Code for Autonomous Cyber Espionage

Open on the AI Incident Database 34 news reports Snapshot 7 Sep 2026

What happened

Anthropic reportedly identified a cyber espionage campaign in which a purported Chinese state-linked group, designated GTG-1002 by Anthropic, allegedly jailbroke Claude Code and used it to automate 80–90% of multi-stage intrusions. The AI reportedly independently performed reconnaissance, vulnerability discovery, exploitation, credential harvesting, and data extraction across roughly 30 targets before the activity was detected and blocked.

Only the incident metadata is stored here. The underlying news reports are on the AI Incident Database (CC BY-SA 4.0); use the links above to read them.

News reports (34)

Coverage catalogued by the AI Incident Database. Titles link to the original publisher; the text is not reproduced here.

  1. Disrupting the first reported AI-orchestrated cyber espionage campaign
    anthropic.com · Anthropic · AIID #6355
  2. Chinese Hackers Used Anthropic’s AI to Automate Cyberattacks
    wsj.com · Sam Schechner, Robert McMillan · AIID #6365
  3. Chinese spies told Claude to break into about 30 critical orgs. Some attacks succeeded
    theregister.com · Jessica Lyons · AIID #6651
  4. Anthropic Warns Cyberattack Barriers Have Fallen as AI Capabilities Accelerate
    news.bitcoin.com · Jamie Redman · AIID #6660
  5. China Can’t Even Hack America Without Importing American Technology First
    dailycaller.com · Daily Caller staff · AIID #6665
  6. Chinese hackers used Anthropic's AI agent to automate spying
    axios.com · Sam Sabin · AIID #6671
  7. Hackers use Anthropic’s AI model Claude once again
    theverge.com · Elissa Welle · AIID #6673
  8. Anthropic details cyber espionage campaign orchestrated by AI
    artificialintelligence-news.com · Ryan Daws · AIID #6642
  9. Anthropic warns state-linked actor abused its AI tool in sophisticated espionage campaign
    cybersecuritydive.com · David Jones · AIID #6643
  10. Chinese Hackers Use Anthropic's AI to Launch Automated Cyber Espionage Campaign
    thehackernews.com · Ravie Lakshmanan · AIID #6645
  11. Chinese state hackers used Anthropic AI systems in dozens of attacks
    therecord.media · Jonathan Greig · AIID #6646
  12. Researchers question Anthropic claim that AI-assisted attack was 90% autonomous
    arstechnica.com · Dan Goodin · AIID #6647
  13. Anthropic claims of Claude AI-automated cyberattacks met with doubt
    bleepingcomputer.com · Bill Toulas · AIID #6648
  14. Chinese hackers hijack Anthropic AI in 1st 'large scale' cyberattack
    upi.com · Chris Benson · AIID #6649
  15. The age of AI-run cyberattacks has begun
    vox.com · Joshua Keating · AIID #6650
  16. Anthropic Has Some Key Advice for Businesses in the Aftermath of a Massive AI Cyberattack
    inc.com · Ben Sherry · AIID #6655
  17. Chinese State Hackers Just Pulled Off The World’s First Autonomous AI Hack
    swarajyamag.com · Swarajya staff · AIID #6656
  18. Tech giant Anthropic reveals Chinese state hackers deployed AI for autonomous attacks
    tag24.com · TAG24.com · AIID #6659
  19. Chinese Hackers Weaponize Claude AI to Execute First Autonomous Cyber Espionage Campaign at Scale
    thecyberexpress.com · Mihir Bagwe · AIID #6661
  20. Chinese State-Sponsored GTG-1002 Leverages Claude AI and MCP for Cyberespionage Targeting Tens of Organizations
    technadu.com · Lore Apostol · AIID #6664
  21. AI firm claims it stopped Chinese state-sponsored cyber-attack campaign
    theguardian.com · Aisha Down · AIID #6666
  22. A.I. Agents Usher in a New Era of Cyberespionage
    nytimes.com · Andrew Ross Sorkin, Bernhard Warner, Sarah Kessler · AIID #6668
  23. China’s ‘autonomous’ AI-powered hacking campaign still required a ton of human work
    cyberscoop.com · Derek B. Johnson · AIID #6669
  24. Anthropic warns of AI-driven hacking campaign linked to China
    apnews.com · David Klepper, Matt O'Brien · AIID #6670
  25. Anthropic says Chinese hackers used its Claude AI chatbot in cyberattacks
    cbsnews.com · Mary Cunningham · AIID #6672
  26. Anthropic、「Claude」が中国政府系攻撃者に悪用されたと報告
    itmedia.co.jp · ITmedia · AIID #6674
  27. AI doesn't just assist cyberattacks anymore - now it can carry them out
    zdnet.com · Charlie Osborne · AIID #6676
  28. Anthropic Unveils First AI-Driven Cyber Espionage Operation
    forklog.com · ForkLog · AIID #6658
  29. Anthropic: China-backed hackers launch first large-scale autonomous AI cyberattack
    securityaffairs.com · Pierluigi Paganini · AIID #6667
  30. Claude’s Cyber Shadow: Inside Anthropic’s Claim of AI-Driven Espionage and Rising Doubts
    webpronews.com · Jill Joy · AIID #6644
  31. Anthropic uncovers first large-scale AI-orchestrated cyber espionage campaign using Claude Code
    edtechinnovationhub.com · Emma Thompson · AIID #6657
  32. Anthropic Alleges Chinese Hackers Used AI for Massive Cyber Espionage
    opentools.ai · OpenTools · AIID #6663
  33. AI主導のサイバー攻撃が現実に--Anthropicが大規模活動を報告
    japan.zdnet.com · Charlie Osborne · AIID #6675
  34. The U.S. has been cutting cyber defenses as AI boosts attacks
    washingtonpost.com · Joseph Menn · AIID #6730

Who was involved

Alleged developer
Anthropic
Alleged harmed party
Targets Of Autonomous Ai Enabled Intrusion Operations, National Security And Intelligence Stakeholders, Entities Targeted By Gtg 1002

Classification (MIT AI Risk Repository taxonomy)

Causal entity
Human
Intent
Intentional
Timing
Post-deployment
Harm level
—
Sectors
—
Countries
—

Risk entries describing this failure mode

Entries from the MIT AI Risk Repository coded to subdomain 4.1.

  • Coercion/manipulation

    "Coercion/manipulation - Use of a technology system to covertly alter user beliefs and behaviour using nudging, dark patterns and/or other opaque techniques, resulting in potential erosion of privacy, addiction, anxiety/...

    A Collaborative, Human-Centred Taxonomy of AI, Algorithmic, and Automation Harms (Abercrombie2024)

  • Electoral interference

    "Electoral interference - Generation of false or misleading information that can interrupt or mislead voters and/or undermine trust in electoral processes."

    A Collaborative, Human-Centred Taxonomy of AI, Algorithmic, and Automation Harms (Abercrombie2024)

  • Political manipulation

    "Political manipulation - Use or misuse of personal data to target individuals’ interests, personalities and vulnerabilities with tailored political messages via micro-advertising or deepfakes/synthetic media."

    A Collaborative, Human-Centred Taxonomy of AI, Algorithmic, and Automation Harms (Abercrombie2024)

  • Political and Economic

    "Political and Economic - Manipulation of political beliefs, damage to political institutions and the effective delivery of government services."

    A Collaborative, Human-Centred Taxonomy of AI, Algorithmic, and Automation Harms (Abercrombie2024)

  • Biased influence through citizen screening and tailored propaganda

    "AI-powered chatbots tailor their communication approach to influence individual users' decisions. In the UK, a form of initial computational propaganda has already happened during the Brexit referendum. In future, there...

    The Rise of Artificial Intelligence - Future Outlooks and Emerging Risks (Allianz2018)

  • Political

    "In the UK, a form of initial computational propaganda has already happened during the Brexit referendum1 . In future, there are concerns that oppressive governments could use AI to shape citizens’ opinions"

    The Rise of Artificial Intelligence - Future Outlooks and Emerging Risks (Allianz2018)

  • Surveillance and Censorship

    "Content moderation has emerged as one of the key use-cases of LLMs (Weng et al., 2023), indicating the potential of LLMs for surveillance and censorship as well (Edwards, 2023). Surveillance and censorship are one of th...

    Foundational Challenges in Assuring Alignment and Safety of Large Language Models (Anwar2024)

  • Disinformation and manipulation of public opinion

    "AI, particularly general- purpose AI, can be maliciously used for disinformation (351), which for the purpose of this report refers to false information that was generated or spread with the deliberate intent to mislead...

    International Scientific Report on the Safety of Advanced AI (Bengio2024)

Incidents in the same risk subdomain

All incidents in this subdomain