The Hacker News | #1 Trusted Source for Cybersecurity News — Index Page

Details have emerged about a now-patched vulnerability in Microsoft 365 Copilot that could enable the theft of sensitive user information using a technique called ASCII smuggling. " ASCII Smuggling is a novel technique that uses special Unicode characters that mirror ASCII but are actually not visible in the user interface," security researcher Johann Rehberger said . "This means that an attacker can have the [large language model] render, to the user, invisible data, and embed them within clickable hyperlinks. This technique basically stages the data for exfiltration!" The entire attack strings together a number of attack methods to fashion them into a reliable exploit chain. This includes the following steps - Trigger prompt injection via malicious content concealed in a document shared on the chat to seize control of the chatbot Using a prompt injection payload to instruct Copilot to search for more emails and documents, a technique called automatic too...

Google has revealed that a security flaw that was patched as part of a software update rolled out last week to its Chrome browser has come under active exploitation in the wild. Tracked as CVE-2024-7965 , the vulnerability has been described as an inappropriate implementation bug in the V8 JavaScript and WebAssembly engine. "Inappropriate implementation in V8 in Google Chrome prior to 128.0.6613.84 allowed a remote attacker to potentially exploit heap corruption via a crafted HTML page," according to a description of the bug in the NIST National Vulnerability Database (NVD). A security researcher who goes by the online pseudonym TheDog has been credited with discovering and reporting the flaw on July 30, 2024, earning them a bug bounty of $11,000. Additional specifics about the nature of the attacks exploiting the flaw or the identity of the threat actors that may be utilizing it have not been released. The tech giant, however, acknowledged that it's aware of the ...

SonicWall has released security updates to address a critical flaw impacting its firewalls that, if successfully exploited, could grant malicious actors unauthorized access to the devices. The vulnerability, tracked as CVE-2024-40766 (CVSS score: 9.3), has been described as an improper access control bug. "An improper access control vulnerability has been identified in the SonicWall SonicOS management access, potentially leading to unauthorized resource access and in specific conditions, causing the firewall to crash," the company said in an advisory released last week. "This issue affects SonicWall Firewall Gen 5 and Gen 6 devices, as well as Gen 7 devices running SonicOS 7.0.1-5035 and older versions." The issue has been addressed in the below versions - SOHO (Gen 5 Firewalls) - 5.9.2.14-13o Gen 6 Firewalls - 6.5.2.8-2n (for SM9800, NSsp 12400, and NSsp 12800) and 6.5.4.15.116n (for other Gen 6 Firewall appliances) SonicWall said the vulnerability is ...

The Dutch Data Protection Authority (DPA) has fined Uber a record €290 million ($324 million) for allegedly failing to comply with European Union (E.U.) data protection standards when sending sensitive driver data to the U.S. "The Dutch DPA found that Uber transferred personal data of European taxi drivers to the United States (U.S.) and failed to appropriately safeguard the data with regard to these transfers," the agency said . The data protection watchdog said the move constitutes a "serious" violation of the General Data Protection Regulation (GDPR). In response, the ride-hailing, courier, and food delivery service has ended the practice. Uber is believed to have collected drivers' sensitive information and retained it on U.S.-based servers for over two years. This included account details and taxi licenses, location data, photos, payment details, and identity documents. In some cases, it also contained criminal and medical data of drivers. The DPA accu...

Nowadays, sensitive and critical data is traveling in everyday business channels that offer only the basic level of security and encryption, and companies are often oblivious to the risk. A case in point: Disney suffered a devastating data leak by a hacktivist group known as NullBulge that got hold of over 1.2 terabytes of data from Disney's internal Slack messaging channels. The breach exposed sensitive information, including: details about unreleased projects, computer code, login details and passwords, and Intellectual Property (IP) and corporate secrets. Slack breaches have also impacted companies like Uber, Rockstar, and Electronic Arts (EA). Cisco Webex used by the German Bundeswehr leaked data from hundreds of meetings, some classified. Outlook was breached by Chinese hackers last year. We have nothing against any of the tools above. They are all great collaboration tools. However, just like companies don’t allow developers to use just any old tool to push code to p...

Cybersecurity researchers are warning about the security risks in the machine learning (ML) software supply chain following the discovery of more than 20 vulnerabilities that could be exploited to target MLOps platforms. These vulnerabilities, which are described as inherent- and implementation-based flaws, could have severe consequences, ranging from arbitrary code execution to loading malicious datasets. MLOps platforms offer the ability to design and execute an ML model pipeline, with a model registry acting as a repository used to store and version-trained ML models. These models can then be embedded within an application or allow other clients to query them using an API (aka model-as-a-service). "Inherent vulnerabilities are vulnerabilities that are caused by the underlying formats and processes used in the target technology," JFrog researchers said in a detailed report. Some examples of inherent vulnerabilities include abusing ML models to run code of the attacker...