Anthropic a introdus noi măsuri împotriva copierii modelelor IA. Rezultate majore în combaterea atacurilor

Anthropic claims to have reduced data extraction from its AI models by up to 99% through new security measures. The company has blocked tens of thousands of accounts amid rising tech competition between the US and China. These measures limit access to Claude's reasoning processes, significantly decreasing the ability of rivals to extract information. Anthropic's security system is said to be robust, with a low false positive rate. The company has identified a Chinese startup, Z.ai, as an example of a competitor attempting to replicate Claude's capabilities.