[untitled]1 points·by thehacknews·5 ay önce·0 comments1 commentsPost comment[–]thehacknews·5 ay öncereplyAnthropic reveals large-scale distillation attacks targeting Claude.Millions of structured prompts were used to extract reasoning, coding, and agentic capabilities.AI model extraction is now a serious cybersecurity and national security threat.
Millions of structured prompts were used to extract reasoning, coding, and agentic capabilities.
AI model extraction is now a serious cybersecurity and national security threat.