Anthropic says China, Iran used its AI to aid spying
Anthropic said it shut down multiple cases of state-sponsored surveillance operations that used its artificial intelligence (AI) models, and warned that governments and state-aligned actors are increasingly using AI to spy on ethnic minorities and dissidents.
Anthropic disclosed that state-sponsored surveillance operations, utilizing its AI models, were identified and dismantled between January and July. These incidents, originating from China, Iran, and West Africa, targeted ethnic minorities and dissidents, including pro-democracy figures in Hong Kong, Tibetan and Falun Gong communities, and Iranian minority groups.
One case involved Iranian actors devising a method to identify individuals via social media accounts, while another saw a contractor for Malian national security authorities use Anthropic's AI, Claude, to develop the software for intelligence gathering. The report also revealed attempts to exploit AI for weapon design, biological research, and dating scams.
Anthropic accused Chinese developers of misleadingly using Claude to enhance their models through a process known as distillation, where one AI model trains another. Moonshot, a Chinese AI developer, is alleged to have rerouted almost 300,000 user requests to Anthropic through a network of 5,380 fraudulent accounts, potentially exposing sensitive user data.
The identities of the individuals behind the biological weapons research, which involved mosquitoes, birds, and pathogens, remain undisclosed to prevent potential harm.
Written by urgent.news from RTE News's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
