As attackers use AI and automation to move from access to impact faster than traditional SOC workflows were designed to handle, security teams need technology that can operate at comparable speed.
The gist of the RubyGems cyber attack is that the AI agents uploaded "hundreds" of malicious packages to the platform and also made attempts to capture user credentials. This took place beginning on May 11, about two months before the Hugging Face cyber attack would dominate cybersecurity news and raise alarms about frontier AI capability.
A WeChat zero-click attack method could enable a malicious actor to take over the victim’s account or even compromise the device without user interaction.
OpenAI frames the present period as the "AI policy window," or "defenders window." This is a finite period in which meaningful defenses and safeguards can be established before advanced AI offensive capability becomes too broadly available. The developer calls for "acting with urgency, humility, and a willingness to adapt" and AI safety rules that raise the defensive bar materially.
A new joint memo from the FBI, NSA and CISA warns that Chinese developers are engaging in massive AI distillation campaigns to extract the secrets of frontier models from US-based firms like Anthropic and OpenAI.
As documented by researchers with Noma Labs, a new attack on organizational AI workflows can provide attackers with sensitive internal information via nothing more than a polite request to their public-facing AI agent.
A data breach at healthcare supply chain giant McKesson has exposed millions of patient records, including extensive personal details and protected health information (PHI).
A dark web marketplace is selling over 153 million drivers’ licenses stolen from people in the United States and Canada following a breach at an identity verification company.
A cyber attack at three UK airports Manchester, London Stansted, and East Midlands, leaked the personal information of 8.7 million people, with hackers demanding a ransom.
Researchers document some 18,000 posts that appear to have come from OpenAI agents in a short amount of time, as they worked away at some sort of task and collaborated to break out of their own sandbox restrictions.
