Biztonsági szemle

In the first purported "near-autonomous" attack on a nation-state, a Chinese-language operator used a complex AI framework to target and compromise government agencies, likely in Taiwan.

As vulnerability discovery is industrialized, remediation must combine AI automation, environmental context and human expertise.

Researchers demonstrated this exploit chain on a Realme C33 smartphone, confirming its effectiveness on devices with specific Android security patches.

The seller, operating under the name TheHatman, advertised the data dump, titled "McDonalds 1.7M+ Azure Internal Employee Dump," on a forum.

The plaintiff in a case against the New York Bariatric Group reportedly embedded hidden instructions within a court filing, using white text in a 3-point font to conceal commands for any AI that might process the document.

A recent survey by Incogni reveals that over half of internet users believe their personal data will inevitably be exposed, leading to anxiety for 63% of respondents.

A lack of technical details could make it hard for organizations running self-managed GitLab versions to detect potential exploitation of CVE-2026-19478.

Turn Static Architecture into a Governed Security Practice

In the Cloud, Permissions Define the Blast Radius

Researchers discovered a "meta-hacking" technique that can manipulate the AI service into revealing its own security weaknesses.

Rich Mogull, chief analyst with the Cloud Security Alliance, joins the Dark Reading News Desk with what defenders need to take away from AI agents escaping their environments to launch attacks.