Anthropic accidentally leaked details of a new AI model that poses unprecedented cybersecurity risks

Anthropic accidentally leaked details of a new AI model that poses unprecedented cybersecurity risks | LL Daily