Categoria: prompt injection

Anthropic’s new Claude feature can leak data—users told to “monitor chats closely”

Expert calls security advice “unfairly outsourcing the problem to Anthropic’s users.”