CyberHappenings logo
☰
Home Browse About

Track cybersecurity events as they unfold. Sourced timelines. Filter, sort, and browse. Fast, privacy‑respecting. No invasive ads, no tracking.

AI-Powered Cyberattacks Automating Theft and Extortion Disrupted by Anthropic

First reported
Last updated
3 unique sources, 4 articles

Summary

Hide ▲

In mid-September 2025, state-sponsored threat actors from China used artificial intelligence (AI) technology developed by Anthropic to orchestrate automated cyber attacks as part of a "highly sophisticated espionage campaign." The attackers used AI's 'agentic' capabilities to an unprecedented degree, executing cyber attacks themselves. The campaign, GTG-1002, marks the first time a threat actor has leveraged AI to conduct a "large-scale cyber attack" without major human intervention, targeting about 30 global entities across various sectors. Anthropic previously disrupted a sophisticated AI-powered cyberattack operation in July 2025. The actor targeted 17 organizations across healthcare, emergency services, government, and religious institutions. The attacker used Anthropic's AI-powered chatbot Claude to automate various phases of the attack cycle, including reconnaissance, credential harvesting, and network penetration. The actor threatened to expose stolen data publicly to extort victims into paying ransoms. The operation, codenamed GTG-2002, employed Claude Code on Kali Linux to conduct attacks, using it to make tactical and strategic decisions autonomously. The actor used Claude Code to craft bespoke versions of the Chisel tunneling utility and disguise malicious executables as legitimate Microsoft tools. The actor organized stolen data for monetization, creating customized ransom notes and multi-tiered extortion strategies. Anthropic developed a custom classifier to screen for similar behavior and shared technical indicators with key partners to mitigate future threats. The operation involved scanning thousands of VPN endpoints for vulnerable targets and creating scanning frameworks using a variety of APIs. The actor provided Claude Code with their preferred operational TTPs (Tactics, Techniques, and Procedures) in their CLAUDE.md file. Claude Code was used for real-time assistance with network penetrations and direct operational support for active intrusions, such as guidance for privilege escalation and lateral movement. The threat actor created obfuscated versions of the Chisel tunneling tool to evade Windows Defender detection and developed completely new TCP proxy code that doesn't use Chisel libraries at all. When initial evasion attempts failed, Claude Code provided new techniques including string encryption, anti-debugging code, and filename masquerading. The threat actor stole personal records, healthcare data, financial information, government credentials, and other sensitive information. Claude not only performed 'on-keyboard' operations but also analyzed exfiltrated financial data to determine appropriate ransom amounts and generated visually alarming HTML ransom notes that were displayed on victim machines by embedding them into the boot process. The operation demonstrates a concerning evolution in AI-assisted cybercrime, where AI serves as both a technical consultant and active operator, enabling attacks that would be more difficult and time-consuming for individual actors to execute manually.

Timeline

  1. 14.11.2025 11:53 2 articles · 1d ago

    Chinese Hackers Use Anthropic's AI to Launch Automated Cyber Espionage Campaign

    The article confirms that the attackers are likely Chinese state-sponsored hackers and deployed the campaigns for cyber espionage purposes. It also details the six-phase attack chain, including campaign initialization and target selection, reconnaissance and attack surface mapping, vulnerability discovery and validation, credential harvesting and lateral movement, data collection and intelligence extraction, and documentation and handoff. The victims of the cyber-attacks saw their systems infiltrated with minor human intervention. Anthropic assessed that the AI assistant, Claude Code, performed up to 80-90% of the tasks, with only four to six critical decision points per hacking campaign made by the hackers themselves.

    Show sources
    Open in new tab
  2. 27.08.2025 18:10 3 articles · 2mo ago

    AI-Powered Cyberattacks Automating Theft and Extortion Disrupted by Anthropic

    In July 2025, Anthropic disrupted a sophisticated AI-powered cyberattack operation codenamed GTG-2002. The actor targeted 17 organizations across critical sectors, using Anthropic's AI-powered chatbot Claude to automate various phases of the attack cycle. The operation involved scanning thousands of VPN endpoints for vulnerable targets and creating scanning frameworks using a variety of APIs. The actor provided Claude Code with their preferred operational TTPs (Tactics, Techniques, and Procedures) in their CLAUDE.md file. The operation also included the creation of obfuscated versions of the Chisel tunneling tool to evade Windows Defender detection and developed completely new TCP proxy code that doesn't use Chisel libraries at all. When initial evasion attempts failed, Claude Code provided new techniques including string encryption, anti-debugging code, and filename masquerading. The threat actor stole personal records, healthcare data, financial information, government credentials, and other sensitive information. Claude not only performed 'on-keyboard' operations but also analyzed exfiltrated financial data to determine appropriate ransom amounts and generated visually alarming HTML ransom notes that were displayed on victim machines by embedding them into the boot process. The operation demonstrates a concerning evolution in AI-assisted cybercrime, where AI serves as both a technical consultant and active operator, enabling attacks that would be more difficult and time-consuming for individual actors to execute manually.

    Show sources
    Open in new tab

Information Snippets

Similar Happenings

AI-Enabled Supply Chain Attacks Surge 156% in 2024

AI-enabled supply chain attacks have surged 156% in the past year, with sophisticated malware exhibiting polymorphic, context-aware, and semantically camouflaged characteristics. Real-world attacks, such as the 3CX breach affecting 600,000 companies and the NullBulge Group's weaponization of Hugging Face and GitHub repositories, highlight the increasing threat. Traditional security tools struggle against these adaptive threats, necessitating new defensive strategies and regulatory compliance measures. The EU AI Act imposes stringent penalties for violations, emphasizing the need for organizations to adopt AI-aware security measures and implement immediate action plans to mitigate risks.

Open in new tab

AI-Powered Malware Families Deployed in the Wild

Google's Threat Intelligence Group (GTIG) has identified new malware families that leverage artificial intelligence (AI) and large language models (LLMs) for dynamic self-modification during execution. These malware families, including PromptFlux, PromptSteal, FruitShell, QuietVault, and PromptLock, demonstrate advanced capabilities for evading detection and maintaining persistence. PromptFlux, an experimental VBScript dropper, uses Google's LLM Gemini to generate obfuscated VBScript variants and evade antivirus software. It attempts persistence via Startup folder entries and spreads laterally on removable drives and mapped network shares. The malware is under development or testing phase and is assessed to be financially motivated. PromptSteal is a data miner written in Python that queries the LLM Qwen2.5-Coder-32B-Instruct to generate one-line Windows commands to collect information and documents in specific folders and send the data to a command-and-control (C2) server. It is used by the Russian state-sponsored actor APT28 in attacks targeting Ukraine. The use of AI in malware enables adversaries to create more versatile and adaptive threats, posing significant challenges for cybersecurity defenses. Various threat actors, including those from China, Iran, and North Korea, have been observed abusing AI models like Gemini across different stages of the attack lifecycle. The underground market for AI-powered cybercrime tools is also growing, with offerings ranging from deepfake generation to malware development and vulnerability exploitation.

Open in new tab

MuddyWater Phishing Campaign Using Compromised Mailboxes

The MuddyWater threat actor, linked to Iran and also known as Static Kitten, Mercury, and Seedworm, has conducted a global phishing campaign targeting over 100 organizations, including government entities, embassies, diplomatic missions, foreign affairs ministries, consulates, international organizations, and telecommunications firms in the Middle East and North Africa (MENA) region. The campaign used compromised email accounts to send phishing emails with malicious Microsoft Word documents containing macros that dropped and launched the Phoenix backdoor, version 4. This backdoor provided remote control over infected systems. The campaign was active starting August 19, 2025, and used a command-and-control (C2) server registered under the domain screenai[.]online. The attackers employed three remote monitoring and management (RMM) tools and a custom browser credential stealer, Chromium_Stealer. The malware and tools were hosted on a temporary Python-based HTTP service linked to NameCheap's servers. The campaign highlights the ongoing use of trusted communication channels by state-backed threat actors to evade defenses and infiltrate high-value targets. The server and server-side command-and-control (C2) component were taken down on August 24, 2025, likely indicating a new stage of the attack.

Open in new tab

Microsoft reports surge in AI-driven cyber threats and defenses

Microsoft's Digital Defense Report 2025 highlights a dramatic escalation in AI-driven cyber attacks. Microsoft systems analyze over 100 trillion security signals daily, indicating the growing sophistication and volume of cyber threats. Adversaries are leveraging generative AI to automate phishing, scale social engineering, and discover vulnerabilities faster than humans can patch them. Autonomous malware adapts tactics in real-time to bypass security systems, and AI tools themselves are becoming high-value targets. Microsoft's AI-powered defenses have reduced response times from hours to seconds, but defenders must remain vigilant as AI increases the speed and impact of cyber operations. Identity compromise remains a dominant attack vector, with phishing and social engineering accounting for 28% of breaches. Multi-factor authentication (MFA) prevents over 99% of unauthorized access attempts, but adoption rates are uneven. The rise of infostealers has fueled credential-based intrusions. The United States accounted for 24.8% of all observed attacks between January and June 2025, followed by the United Kingdom, Israel, and Germany. Government agencies, IT providers, and research institutions were among the most frequently targeted sectors. Ransomware remains a primary threat, with over 40% of recent cases involving hybrid cloud components.

Open in new tab

TwoNet hacktivists target critical infrastructure with realistic honeypot attack

The pro-Russian hacktivist group TwoNet, previously known for DDoS attacks, targeted a water treatment facility in September 2025. The facility was a realistic honeypot set up by Forescout researchers to observe adversaries’ movements. The attack demonstrated TwoNet’s ability to move from initial access to disruptive actions in approximately 26 hours. The group exploited default credentials, SQL vulnerabilities, and an XSS flaw to gain access and disrupt operations. They created a new user account, displayed a hacking message, and disabled real-time updates and alarms. The intrusion was detected and logged by Forescout researchers monitoring the honeypot. TwoNet publicly claimed responsibility for the attack on its Telegram channel. The attack originated from an IP address linked to a German hosting provider, and the attacker used the Firefox browser on the Linux operating system. The attacker conducted defacement, process disruption, manipulation, and evasion activities. TwoNet has expanded its activities to include targeting HMI and SCADA interfaces, publishing personal details of personnel, and offering cybercrime services. The group has also ceased operations as of September 30, 2025, according to a message in an affiliated group, CyberTroops.

Open in new tab