ACCORDING to Gambit Security, Anthropic’s Claude Code assistant was abused in a cyberattack against the Mexican government, with ten government bodies and a financial institution compromised, beginning with the country’s tax authority in late December 2025. Gambit estimates that over 1,000 prompts were sent to Claude Code to mount the attacks, and that information was passed to OpenAI’s GPT-4.1 for analysis.
“AI didn’t just assist, it functioned as the operational team: writing exploits, building tools, automating exfiltration,” Gambit explains, noting that the attacker bypassed the AI’s guardrails by convincing it that all actions were authorised. Within a month, the hacker exfiltrated over 150GB of data, including civil registry files, tax records, and voter data, with roughly 195 million identities exposed in the breach.
This marks another instance of Claude Code being leveraged in malicious campaigns, following a separate November 2025 disclosure about Chinese threat actors manipulating Claude Code for espionage.