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 researcher discloses four IBM zero-days after company refused to patch
April 21, 2020
A security researcher has published today details about four zero-day vulnerabilities impacting an IBM security product after the company refused to patch bugs following a private bug disclosure attempt. The bugs impact the IBM Data Risk Manager (IDRM), an enterprise security tool that aggregates feeds from vulnerability scanning tools and other risk management tools to let admins ...
- Many problems with cyber security of Schipihol’s border control: Court of Audit
April 20, 2020
Schiphol is very vulnerable to cyber attacks, the Court of Audit concluded after investigating the cyber security of the border control systems the Koninklijke Marechaussee uses at the airport. Two of the three systems are not properly protected against cyber attacks, NOS reports. Systems at Schiphol are hardly ever tested for how well they can stand ...
- Starbleed bug impacts FPGA chips used in data centers, IoT devices, industrial equipment
April 20, 2020
A team of academics says they’ve discovered a new security bug that impacts Xilinx FPGA (Field Programmable Gate Arrays) chipsets. Named Starbleed, the bug allows attackers — with both physical or remote access — to extract and tamper with an FGPA’s bitstream (configuration file) to reprogram the chip with malicious code. FPGAs are add-in cards that can ...
- DHS CISA: Companies are getting hacked even after patching Pulse Secure VPNs
April 17, 2020
Companies that run Pulse Secure VPN servers are still at risk of getting hacked, despite patching vulnerable systems, cyber-security agencies from the US and Japan have warned this month. Pulse Secure VPN servers are enterprise-grade VPN gateways that companies use to let workers connect to internal company networks from across the internet. Last year, a major vulnerability ...
- Academics steal data from air-gapped systems using PC fan vibrations
April 17, 2020
Academics from an Israeli university have proven the feasibility of using fans installed inside a computer to create controlled vibrations that can be used to steal data from air-gapped systems. The technique, codenamed AiR-ViBeR, is the latest in a long list of wacky data exfiltration techniques devised by Mordechai Guri, the head of R&D at the ...
- Apple Safari Flaws Enable One-Click Webcam Access
April 6, 2020
A security researcher has disclosed vulnerabilities in Apple’s Safari browser that can be used to snoop on iPhones, iPads and Mac computers using their microphones and cameras. To exploit the flaws in a real-world attack, all an attacker would need to do is convince a victim to click one malicious link. Security researcher Ryan Pickren has revealed ...