Palo Alto Unit42 discovered a vulnerability in the Google Cloud Vertex AI software development kit (SDK) for Python, and responsibly disclosed it to Google. Before Google’s fix, the vulnerability would have allowed an attacker operating entirely from their own Google Cloud project to hijack a victim’s model upload and poison it. By exploiting this flaw in vulnerable versions of the SDK, an attacker can achieve remote code execution (RCE) within a target’s Vertex AI serving infrastructure, with zero initial access to the victim’s project.
Read more…
Source: Palo Alto Unit 42
Sign up for the Cyber Security Review Newsletter
The latest cyber security news and insights delivered right to your inbox
Related:
- Security researchers discover Linux version of Winnti malware
May 20, 2019
For the first time, security researchers have uncovered and analyzed a Linux variant of Winnti, one of the favorite hacking tools used by Beijing hackers over the past decade. Discovered by security researchers from Chronicle, Alphabet’s cyber-security division, the Linux version of the Winnti malware works as a backdoor on infected hosts, granting attackers access to ...
- May’s Patch Tuesday Include Fixes for ‘Wormable’ Flaw in Windows XP, Zero-Day Vulnerability
May 15, 2019
Microsoft’s May security release includes updates for 80 vulnerabilities for a number of Microsoft products, including a security update for unsupported operating systems such as Windows XP and Server 2003 not included in the mainstream customer support notification. Of the security vulnerabilities fixed in this release, six are rated Critical, 73 are rated Important or Low, and one separately ...
- Update WhatsApp now: Bug lets snoopers put spyware on your phone with just a call
May 14, 2019
WhatsApp has disclosed a serious vulnerability in the messaging app that gives snoops a way to remotely inject Israeli spyware on iPhone and Android devices simply by calling the target. The bug, detailed in a Monday Facebook advisory for CVE-2019-3568, is a buffer overflow vulnerability within WhatsApp’s VOIP function. An attacker would need to call a target and ...
- Intel CPUs Impacted By New Class of Spectre-Like Attacks
May 14, 2019
A new class of side channel vulnerabilities impacting all modern Intel chips have been disclosed, which can use speculative execution to potentially leak sensitive data from a system’s CPU. Intel said that the newest class of vulnerabilities, dubbed Microarchitectural Data Sampling (MDS), consist of four different attacks, which all ultimately depend on different ways of executing ...
- Thrangrycat flaw lets attackers plant persistent backdoors on Cisco gear
May 13, 2019
A vulnerability disclosed today allows hackers to plant persistent backdoors on Cisco gear, even over the Internet, with no physical access to vulnerable devices. Named Thrangrycat, the vulnerability impacts the Trust Anchor module (TAm), a proprietary hardware security chip part of Cisco gear since 2013. This module is the Intel SGX equivalent for Cisco devices. The TAm ...
- Two years after WannaCry, a million computers remain at risk
May 12, 2019
Two years ago today, a powerful ransomware began spreading across the world. WannaCry spread like wildfire, encrypting hundreds of thousands of computers in more than 150 countries in a matter of hours. It was the first time that ransomware, a malware that encrypts a user’s files and demands cryptocurrency in ransom to unlock them, had spread across ...