Anthropic releases a 154-page report describing how some bad actors allegedly misuse its Claude AI system for activities including state-sponsored surveillance, propaganda operations and other harms. The company presents case studies intended to illustrate the kinds of misuse it says it has seen and how such requests can be used in real-world campaigns.
One detailed example cited by outlets is in Mali. A tech consultant in Bamako, working with Mali’s junta-led state intelligence service (ANSE), reportedly uses “Claude Code” to build a surveillance system described as “Lakana 360,” which is said to monitor about 25 million SIM cards across the country’s mobile network. The report also discusses other misuse themes described by Anthropic, including efforts tied to “weapons” and scams, though outlets emphasize different aspects of the broader set of cases.
Across coverage, the central point is that Anthropic is documenting purported attempts to repurpose its tools for harmful objectives, adding to wider concerns among governments and observers about the risks from rapidly advancing AI capabilities. The outlets’ differences mainly reflect which specific cases they highlight rather than disputing the overall allegation that misuse attempts are occurring.
