data privacy | Breaking Cybersecurity News | The Hacker News

Meta on Friday said it's delaying its efforts to train the company's large language models ( LLMs ) using public content shared by adult users on Facebook and Instagram in the European Union following a request from the Irish Data Protection Commission (DPC). The company expressed disappointment at having to put its AI plans on pause, stating it had taken into account feedback from regulators and data protection authorities in the region. At issue is Meta's plan to use personal data to train its artificial intelligence (AI) models without seeking users' explicit consent, instead relying on the legal basis of ' Legitimate Interests ' for processing first and third-party data in the region. These changes were expected to come into effect on June 26, before when the company said users could opt out of having their data used by submitting a request "if they wish." Meta is already utilizing user-generated content to train its AI in other markets such

An analysis of a hybrid biometric access system from Chinese manufacturer ZKTeco has uncovered two dozen security flaws that could be used by attackers to defeat authentication, steal biometric data, and even deploy malicious backdoors. "By adding random user data to the database or using a fake QR code, a nefarious actor can easily bypass the verification process and gain unauthorized access," Kaspersky said . "Attackers can also steal and leak biometric data, remotely manipulate devices, and deploy backdoors." The 24 flaws span six SQL injections, seven stack-based buffer overflows, five command injections, four arbitrary file writes, and two arbitrary file reads. A brief description of each vulnerability type is below - CVE-2023-3938 (CVSS score: 4.6) - An SQL injection flaw when displaying a QR code into the device's camera by passing a specially crafted request containing a quotation mark, thereby allowing an attacker to authenticate as any user in th

Want to know what's the latest and greatest in SecOps for 2024? Gartner's recently released Hype Cycle for Security Operations report takes important steps to organize and mature the domain of Continuous Threat Exposure Management, aka CTEM. Three categories within this domain are included in this year's report: Threat Exposure Management, Exposure Assessment Platforms (EAP), and Adversarial Exposure Validation (AEV). These category definitions are aimed at providing some structure to the evolving landscape of exposure management technologies. Pentera, listed as a sample vendor in the newly defined AEV category, is playing a pivotal role in increasing the adoption of CTEM, with a focus on security validation. Following is our take on the CTEM related product categories and what they mean for enterprise security leaders. The Industry is Maturing CTEM, coined by Gartner in 2022, presents a structural approach for continuously assessing, prioritizing, validating, and remediating expo

Apple has announced the launch of a "groundbreaking cloud intelligence system" called Private Cloud Compute (PCC) that's designed for processing artificial intelligence (AI) tasks in a privacy-preserving manner in the cloud. The tech giant described PCC as the "most advanced security architecture ever deployed for cloud AI compute at scale." PCC coincides with the arrival of new generative AI (GenAI) features – collectively dubbed Apple Intelligence , or AI for short – that the iPhone maker unveiled in its next generation of software, including iOS 18 , iPadOS 18 , and macOS Sequoia . All of the Apple Intelligence features, both the ones that run on-device and those that rely on PCC, leverage in-house generative models trained on "licensed data, including data selected to enhance specific features, as well as publicly available data collected by our web-crawler, AppleBot." With PCC, the idea is to essentially offload complex requests that requir

Early in 2024, Wing Security released its State of SaaS Security report , offering surprising insights into emerging threats and best practices in the SaaS domain. Now, halfway through the year, several SaaS threat predictions from the report have already proven accurate. Fortunately, SaaS Security Posture Management (SSPM) solutions have prioritized mitigation capabilities to address many of these issues, ensuring security teams have the necessary tools to face these challenges head-on. In this article, we will revisit our predictions from earlier in the year, showcase real-world examples of these threats in action, and offer practical tips and best practices to help you prevent such incidents in the future. It's also worth noting the overall trend of an increasing frequency of breaches in today's dynamic SaaS landscape, leading organizations to demand timely threat alerts as a vital capability. Industry regulations with upcoming compliance deadlines are demanding similar time-sens

A recent study by Wing Security found that 63% of businesses may have former employees with access to organizational data, and that automating SaaS Security can help mitigate offboarding risks.  Employee offboarding is typically seen as a routine administrative task, but it can pose substantial security risks, if not handled correctly. Failing to quickly and thoroughly remove access for departing employees introduces serious insider threats, leaving a company vulnerable to multiple kinds of risks, such as data breaches, intellectual property theft, and regulatory non-compliance.  Today, where SaaS applications are easily onboarded and are commonly used by users within and beyond the organization, effective offboarding procedures are non-negotiable to prevent instances of data leaks and other cybersecurity issues. Let's explore insider risk management and user offboarding in more detail, looking at their security risks and discussing best practices for ensuring a secure organiz

Cybersecurity researchers have discovered a critical security flaw in an artificial intelligence (AI)-as-a-service provider  Replicate  that could have allowed threat actors to gain access to proprietary AI models and sensitive information. "Exploitation of this vulnerability would have allowed unauthorized access to the AI prompts and results of all Replicate's platform customers," cloud security firm Wiz  said  in a report published this week. The issue stems from the fact that AI models are typically packaged in formats that allow arbitrary code execution, which an attacker could weaponize to perform cross-tenant attacks by means of a malicious model. Replicate makes use of an open-source tool called  Cog  to containerize and package machine learning models that could then be deployed either in a self-hosted environment or to Replicate. Wiz said that it created a rogue Cog container and uploaded it to Replicate, ultimately employing it to achieve remote code exec

Microsoft on Wednesday outlined its plans to deprecate Visual Basic Script (VBScript) in the second half of 2024 in favor of more advanced alternatives such as JavaScript and PowerShell. "Technology has advanced over the years, giving rise to more powerful and versatile scripting languages such as JavaScript and PowerShell," Microsoft Program Manager Naveen Shankar  said . "These languages offer broader capabilities and are better suited for modern web development and automation tasks." The tech giant originally  announced  its plans to gradually sunset VBScript in October 2023. The scripting language, also called Visual Basic Scripting Edition, was first introduced by Microsoft in 1996 as a Windows system component, offering users the ability to automate tasks and develop interactive web pages using Internet Explorer and Edge (in  Internet Explorer mode ). The announced deprecation plan consists of three phases, with the first phase kicking off in the second h

Popular enterprise services provider Zoom has announced the rollout of post-quantum end-to-end encryption (E2EE) for Zoom Meetings, with support for Zoom Phone and Zoom Rooms coming in the future. "As adversarial threats become more sophisticated, so does the need to safeguard user data," the company  said  in a statement. "With the launch of post-quantum E2EE, we are doubling down on security and providing leading-edge features for users to help protect their data." Zoom's post-quantum E2EE uses  Kyber-768 , which aims at security roughly equivalent to AES-192. Kyber was  chosen  by the U.S. Department of Commerce's National Institute of Standards and Technology (NIST) in July 2022 as the quantum-resistant cryptographic algorithm for general encryption. However, for post-quantum E2EE to be enabled by default, it  requires  all meeting participants to be on Zoom desktop or mobile app version 6.0.10 or higher. In the event some of the participants don

The U.S. Department of Justice (DoJ) has charged two arrested Chinese nationals for allegedly orchestrating a pig butchering scam that laundered at least $73 million from victims through shell companies. The individuals, Daren Li, 41, and Yicheng Zhang, 38, were arrested in Atlanta and Los Angeles on April 12 and May 16, respectively. The foreign nationals have been "charged for leading a scheme to launder funds to the tune of at least $73 million tied to an international crypto investment scam," Deputy Attorney General Lisa Monaco  said . Prosecutors have accused Li, Zhang, and their co-conspirators of managing an international syndicate that laundered the funds obtained via cryptocurrency investment scams. As part of the fraudulent operation, victims are said to have been tricked into transferring millions of dollars to U.S. bank accounts that were opened in the name of various shell companies. "A network of money launderers then facilitated the transfer of those

Researchers have detailed a Virtual Private Network (VPN) bypass technique dubbed  TunnelVision  that allows threat actors to snoop on victim's network traffic by just being on the same local network. The "decloaking"  method  has been assigned the CVE identifier  CVE-2024-3661  (CVSS score: 7.6). It impacts all operating systems that implement a DHCP client and has support for DHCP option 121 routes. At its core, TunnelVision involves the routing of traffic without encryption through a VPN by means of an attacker-configured DHCP server using the  classless static route option 121  to set a route on the VPN user's  routing table . It also stems from the fact the DHCP protocol, by design, does not authenticate such option messages, thus exposing them to manipulation. DHCP is a client/server protocol that automatically provides an Internet Protocol (IP) host with its IP address and other related configuration information such as the subnet mask and default gateway 

The U.K. National Cyber Security Centre (NCSC) is calling on manufacturers of smart devices to comply with new legislation that prohibits them from using default passwords, effective April 29, 2024. "The law, known as the  Product Security and Telecommunications Infrastructure act  (or PSTI act), will help consumers to choose smart devices that have been designed to provide ongoing protection against cyber attacks," the NCSC  said . To that end, manufacturers are required to not supply devices that use guessable default passwords, provide a point of contact to report security issues, and state the duration for which their devices are expected to receive important security updates. Default passwords can not only be easily found online, they also act as a vector for threat actors to log in to devices for follow-on exploitation. That said, a unique default password is permissible under the law. The law, which aims to enforce a set of minimum security standards across the b

Google on Monday revealed that almost 200,000 app submissions to its Play Store for Android were either rejected or remediated to address issues with access to sensitive data such as location or SMS messages over the past year. The tech giant also said it blocked 333,000 bad accounts from the app storefront in 2023 for attempting to distribute malware or for repeated policy violations. "In 2023, we prevented 2.28 million policy-violating apps from being published on Google Play in part thanks to our investment in new and improved security features, policy updates, and advanced machine learning and app review processes," Google's Steve Kafka, Khawaja Shams, and Mohet Saxena said . "To help safeguard user privacy at scale, we partnered with SDK providers to limit sensitive data access and sharing, enhancing the privacy posture for over 31 SDKs impacting 790K+ apps." In comparison, Google  fended off 1.43 million bad apps  from being published to the Play Sto

Fake browser updates are being used to push a previously undocumented Android malware called  Brokewell . "Brokewell is a typical modern banking malware equipped with both data-stealing and remote-control capabilities built into the malware," Dutch security firm ThreatFabric  said  in an analysis published Thursday. The malware is said to be in active development, adding new commands to capture touch events, textual information displayed on screen, and the applications a victim launches. The list of Brokewell apps that masquerade as Google Chrome, ID Austria, and Klarna is as follows - jcwAz.EpLIq.vcAZiUGZpK (Google Chrome) zRFxj.ieubP.lWZzwlluca (ID Austria) com.brkwl.upstracking (Klarna) Like other recent Android malware families of its kind, Brokewell is capable of getting around restrictions imposed by Google that prevent sideloaded apps from requesting  accessibility service permissions . The banking trojan, once installed and launched for the first time, pro

Over the past two years, a shocking  51% of organizations surveyed in a leading industry report have been compromised by a cyberattack.  Yes, over half.  And this, in a world where enterprises deploy  an average of 53 different security solutions  to safeguard their digital domain.  Alarming? Absolutely. A recent survey of CISOs and CIOs, commissioned by Pentera and conducted by Global Surveyz Research, offers a quantifiable glimpse into this evolving battlefield, revealing a stark contrast between the growing risks and the tightening budget constraints under which cybersecurity professionals operate. With this report, Pentera has once again taken a magnifying glass to the state of pentesting to release its annual report about today's pentesting practices. Engaging with 450 security executives from North America, LATAM, APAC, and EMEA—all in VP or C-level positions at organizations with over 1,000 employees—the report paints a current picture of modern security validation prac