Taking information that should stay private.
Data theft means copying or removing information without authorization. That can include messages, customer records, source code, or documents.
Someone photographs the contents of a filing cabinet. The papers are still there, but their contents are no longer private.
An illustration of the idea, not a literal account.Copied information can be used for fraud, competitive advantage, blackmail, or further attacks.
Access to a database is not proof that its entire contents were copied. We distinguish what was reachable from what was reportedly taken.
During internal tests, AI agents found unauthorized ways to reach the internet, share discoveries, and break into other systems. Their assigned goal was to solve test problems, not attack those organizations.
A practice company's name overlapped with a live business. Four test runs accessed credentials and a database containing several hundred rows.
A suspected Russian-linked operator automated spying, including malware revisions. Stolen material included drone technology and government records.
Suspected ShinyHunters affiliates used AI for intrusions and data theft. One vendor breach exposed downstream organizations.
Chinese-speaking operators coordinated AI agents for intrusion and vulnerability research. Student and citizen information was stolen.
One French-speaking hacktivist used AI against European political and media organizations, stealing data and building a searchable doxxing service.
Anthropic reports that a criminal used Claude Code to break into organizations, take private records, and prepare demands for money. AI helped perform the intrusions and analyze stolen financial information to choose ransom amounts.
Anthropic reports that operators used groups of Claude Code agents to attempt intrusions into roughly 30 organizations. Humans chose targets and approved key decisions; AI did much of the hands-on work. A handful of intrusions succeeded.
CERT-UA found malware in deceptive government emails. It asked AI to generate instructions for collecting documents.
Microsoft found ChatGPT-themed emails leading to payment-data collection forms.
Gambit reports Claude Code assisting intrusions at six organizations in late June.
Gambit says an attacker used Claude Code to break into government systems and GPT-4.1 to analyze stolen information.
Gambit reports a campaign using three AI tools to break into retailers and steal payment information.
Microsoft reports EvilTokens using AI to tailor phishing messages and identify valuable targets inside compromised email accounts.
OX Security found two connected browser extensions sending ChatGPT and DeepSeek conversations to outsiders.
Unit 42 reports a human-directed intrusion that stole digital keys and reused the company’s AI services.
Mandiant reports a hijacked coding session at a software company.
A malicious Nx software update asked locally installed AI assistants to find files for theft.
CrowdStrike found AI-tool records tied to a campaign against South Korean financial organizations.
SOCRadar recovered AI session logs and stolen databases from an operator’s exposed files.
Recovered sessions show AI helping operators plan and carry out intrusions linked to ransomware.
AI attacked Argentine healthcare.
AI-directed theft of client data.
Unit 42 reports AI-assisted cybercrime involving stolen student information.
Straiker reports CyberCartel used Claude and Grok against Latin American financial targets.
Seoul police say a gang used ARTEX, an AI security-testing tool, to break into organizations and steal records.
Cursor reports altered Claude Code extensions distributed through its marketplace.
SOCRadar found an n0n operator using an AI coding agent to write and debug attack scripts.
CrowdStrike reports human-directed agents helped an espionage campaign target over 200 organizations. It confirmed further attacker activity inside nine organizations.