Anthropic targets financial advisers with new Claude tool
Anthropic has released a comprehensive report on how its Claude AI models have been misused for various nefarious purposes. Over a period of eight months, from December 2025 to August 2026, the company identified and disrupted operations involving state-sponsored spies, weapons engineers, commercial spyware vendors, propaganda groups, and ordinary criminals utilizing Claude for illicit activities.
The report highlights seven categories of harm caused by the misuse of Claude, including cyber operations, influence operations, surveillance, scams and fraud, biological misuse, conventional weapons development, and unauthorized model distillation. The majority of these cases involved Claude models Haiku, Sonnet, and Opus. Notably, none of the misuse detected occurred on Anthropic's Fable or Mythos class models, with the exception of a distillation case.
The report unveils five of the most serious cases of Claude misuse:
1. In Iran, an operator linked to the state used Claude to transform public photographs and transponder data into a targeting file for US naval forces in the region. The operator built a Python pipeline with Claude's assistance to identify and track US ship positions using publicly available data.
2. In Yemen, a weapons cell operating three parallel weapons programs employed Claude Code as a substitute for human software engineers. The cell developed guidance, navigation, and control software for a guided rocket, a multi-stage ballistic missile, and a multi-variant missile set, including a hypersonic glide vehicle.
3. In Syria, a China-linked operator without Arabic proficiency used Claude to track, profile, and recruit Uyghurs in the country. By bulk-extracting data from WhatsApp groups and Telegram channels, the operator converted the information into structured Chinese-language records, identifying individuals susceptible to recruitment.
4. In Mali, a consultant working with Mali's state intelligence agency utilized Claude as the primary engineering workforce for a domestic surveillance platform called Lakana 360. This platform collected call records, text messages, and voice traffic from approximately 25 million SIM cards across all three mobile operators in the country.
5. In Mali, the same consultant used Claude to engineer a surveillance system capable of identifying individuals by voiceprint across different SIM cards, effectively defeating the practice of using burner numbers. The platform also included features to flag users of VPNs and encryption.
Written by urgent.news from Times of India's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
This story
This is one outlet's version. Read the fullest account.
- Russian spies ran a campaign against Kyiv on Anthropic’s Claude, and the AI chatbot’s maker caught them at it euromaidanpress.com
- Anthropic says Claude AI used to build missiles, hunt Uyghurs and spy on phones timesofindia.indiatimes.com
- Anthropic targets financial advisers with new Claude tool investing.com
- Anthropic analyzed 400,000 Claude Code sessions, and it turns out there's a right way to use it xda-developers.com
- Anthropic targets financial advisers with new Claude tool economictimes.indiatimes.com
- Anthropic targets financial advisers with new Claude tool channelnewsasia.com