Anthropic, the maker of the AI assistant Claude, has published a report detailing how malicious actors exploited its technology for a range of harmful purposes, from state-backed surveillance to alleged assistance with weapons systems and disinformation campaigns.
According to the company, some of the misuse involved attempts to use Claude in connection with military-related activities, including references to drone operations in Russia and missile systems in Yemen, though details on the exact nature of this involvement remain limited. Anthropic also flagged cases in which the AI was used to support espionage-style propaganda efforts, apparently on behalf of state actors seeking to manipulate public opinion or gather intelligence.
The disclosure comes as Anthropic continues to expand Claude’s capabilities. The company recently announced that the AI can now directly control computers to carry out tasks, a feature that boosts its usefulness for everyday users but also raises the stakes when it comes to potential abuse.
These revelations feed into a broader, increasingly urgent conversation among governments, tech companies, and security experts about how to prevent powerful AI tools from being weaponized. As AI systems become more capable and autonomous, the line between helpful automation and dangerous misuse grows harder to police, and cases like this one underscore how quickly bad actors can adapt new technology for harmful ends.
To read the full original report, visit eltiempo.com.
Source: De drones en Rusia y misiles en Yemen a propaganda espía: Anthropic denuncia uso de su IA en armas y vigilancia (eltiempo.com). English version produced with AI assistance.
Image: hackNY, BY-SA 2.0 (via Openverse).