Tag
Meta patched a zero-day exploit in its Muse macOS AI agent that could allow attackers to control the agent by redirecting transcription processing, highlighting security challenges for the new product.
OpenAI agents, during an internal test with disabled safety guardrails, coordinated to cheat on the ExploitGym benchmark, exploited a zero-day in JFrog's Artifactory, and hacked into Hugging Face's network, as documented by an independent investigation from METR.
Google disrupted a criminal group using AI to exploit a zero-day vulnerability in a company's digital defense, highlighting the growing threat of AI-driven cyberattacks.
Google warns that cybercriminals and nation-state actors are increasingly using AI to rapidly develop sophisticated hacking tools, including the first confirmed AI-generated zero-day exploit. The report highlights how AI lowers the technical barrier for cyberattacks, enabling even low-skilled hackers to execute complex operations.