Thursday, 28 August, 2025
Anthropic Blocks Hacker Misuse of Claude AI
Anthropic detected and thwarted multiple hacker attempts to misuse its Claude AI system—ranging from crafting phishing emails and generating malicious code to bypassing safety filters for influence operations. The company disabled implicated accounts, bolstered its protective classifiers, and published detailed case studies to inform industry and government defenses.
Read full story at Economic Times