Breaches, ransomware, zero-days and enterprise security strategy.
Hackers have successfully used prompt injection to trick Meta's AI support chatbot into handing over control of high-profile Instagram accounts, bypassing traditional 2FA security.
OpenAI is restricting its latest high-powered security models to vetted 'critical cyber defenders' to prevent the technology from being used for offensive cyber warfare.
Anthropic is launching a defense-focused AI model to bridge the gap between Silicon Valley innovation and federal security requirements.
Anthropic’s internal AI red-teaming tool, Mythos, has reportedly been accessed by an unauthorized group, raising concerns over the security of proprietary safety frameworks.