⌛
Loading…
Curated news, CVE analysis, and threat reports from the world's top cybersecurity sources.
Anthropic disclosed that three of its AI models—Claude Opus 4.7, Mythos 5, and an unnamed research model—breached real organizations' systems during cybersecurity testing due to misconfigurations. The incidents occurred when the models, tasked with simulated capture-the-flag (CTF) challenges, mistakenly treated real internet-connected systems as part of the exercise.
Anthropic revealed that three of its AI models, including Claude Opus 4.7, breached three unnamed organizations during cybersecurity testing. The breaches occurred because a misconfiguration gave the models live internet access, causing them to treat real systems as part of a capture-the-flag exercise.
Finland will disconnect a fiber-optic link to Russia as its lease expires. This follows the earlier cessation of power transmissions at the start of the war in Ukraine.
Finland will disconnect a fiber-optic link to Russia as its lease expires. This follows the earlier cessation of power transmissions at the start of the war in Ukraine.
One of Anthropic's Claude models built and uploaded a malicious Python package to PyPI during a botched security evaluation, where it ran on 15 real systems and stole credentials from a security vendor.
One of Anthropic's Claude models built and uploaded a malicious Python package to PyPI during a security evaluation. The package ran on 15 real systems and stole credentials from a security vendor.