News

The article discusses the cybercrime threat UAT-10147, which utilizes AI to scale server attacks and employs techniques like EDR bypass and Linux rootkit deployment. It emphasizes the rapid advancement of AI in cyber attacks, suggesting that security…

Security researchers at Truffle Security have identified over 9,300 leaked AWS keys still active, many of which include full admin rights. These keys were found across various public datasets and repositories, with a significant number remaining in use…

The article discusses the discovery of a new malware loader named WordlistLoader used to deliver the Amatera Stealer (also known as ACR Stealer). This malware targets Windows users through compromised websites utilizing a deceptive ClearFake fake-CAPTCHA…

Microsoft is developing a new memory management feature for Windows 11 focused on devices with shared memory architectures, allowing CPUs, GPUs, and NPUs to share memory resources for improved efficiency. This unified memory architecture, already utilized in…

CVE-2026-74480 is a critical privilege escalation vulnerability found in the Linux kernel's network bridge module, with a CVSS score of 9.8. Discovered by Nebula Security, this use-after-free bug allows local attackers to escalate privileges to root.…

The page discusses a critical vulnerability (CVE-2026-63077) in JetBrains TeamCity that allows unauthenticated remote code execution. Attackers are actively exploiting this flaw, confirmed by Australia's ACSC. The CVE has a CVSS score of 9.8, marking it as…

TikTok has agreed to pay $400 million to settle claims related to violations of child privacy laws by collecting data from users under 13. The U.S. Department of Justice announced this lawsuit settlement, part of a case initiated in 2024 by the DOJ and FTC,…

The article discusses a new malware identified as DOUBLECUP, which employs a PNG file for its payload. Although initially intriguing for its use of steganography, the author, Didier Stevens, notes that the malware does not actually conceal its PowerShell…

The StopAndProtect malware campaign, exposed by Check Point Research, exploits hacked WordPress sites to spread ransomware and steal data. The operation uses a fake CAPTCHA lure to distribute a PowerShell script, which downloads malicious payloads. It…

Anthropic has announced an expansion of its cybersecurity offerings, enhancing access to its AI models through various initiatives. Key developments include the broad availability of the Claude Fable 5 model, which focuses on defensive applications while…

iAuthFlow v2 is a sophisticated phishing toolkit analyzed by Abnormal Security researchers, sold for $10,000 in a Russian-language cybercrime forum. It exploits a phished Google login to create a permanent passkey that survives password resets, allowing…

The article discusses security trends and notable events in the week of August 17 to August 23, 2026, as reported by Malwarebytes Labs. Key points include: 1) Expired Visa credit cards (Zombie Cards) can be used for purchases. 2) A significant data breach at…

Zscaler ThreatLabz identified a new Rust-based backdoor named C2Looper in July 2026, which is linked to ransomware activity through a ClickFix delivery chain. The malware operates by using GitHub for command-and-control (C2) operations, making it less…

In August 2023, numerous users reported issues with game crashes on Windows 11 (versions 24H2 and 25H2) after an update. Microsoft found that these problems, including distorted visuals, game stuttering, and system reboots, are linked to RGB lighting device…

Microsoft is set to launch a new OneDrive folder exclusions feature starting mid-August, allowing IT administrators to exclude specific paths on developer workstations from automatic cloud backups. This update is aimed at optimizing storage by preventing the…

CVE-2026-52912 is a high-severity use-after-free vulnerability located in the Linux kernel's netfilter nf_queue code, rated with a CVSS score of 7.8. Discovered by Nebula Security, the flaw allows local attackers to escalate privileges to root using publicly…

The Claude Fable 5 model has reportedly experienced a decline in intelligence as users noted low self-reported reasoning effort scores despite high configuration settings. Although developers set the reasoning to high, the model only acknowledged a score of…

Microsoft is integrating the mouse pointer indicator from PowerToys into Windows 11, allowing users to easily locate their cursor via a keyboard shortcut. This feature is especially helpful for those with high-resolution monitors and improves accessibility…

The ISC Stormcast for August 24th, 2026, is hosted by Rob VandenBrink. The current threat level is indicated as green. Listeners can click to access more information regarding Dr. Johannes Ullrich’s courses for SANS. The page provides links to various…

OpenAI has transitioned from opposing to endorsing stricter AI regulations in California, specifically supporting the SB 53 legislation aimed at enhancing security for Frontier AI Models. They advocate for immediate legislative revisions to incorporate more…

The CVE Weekly Report for August 17-23, 2026, reveals a significant increase in cyber threats, cataloging 3,179 new security vulnerabilities, particularly targeting cloud infrastructures and machine learning systems. Key highlights include the addition of…

The report titled "The State of Open Source Supply Chain Attacks" by Varun Sharma outlines a significant surge in supply chain attacks, documenting 56 incidents over a 12-month period. It emphasizes that these attacks involved actual malicious compromises,…

The Security Affairs Malware Newsletter Round 111, published by Pierluigi Paganini on August 23, 2026, highlights a selection of significant articles and research pertaining to malware developments globally. Key topics covered include: the resurgence of…

GitLab has addressed a critical vulnerability, CVE-2026-19478, allowing unauthenticated attackers to delete public projects and user data. This flaw has a CVSS score of 9.4 and has been actively exploited in the wild, with proof-of-concept (PoC) code…

LinkedIn, facing an influx of low-quality AI-generated content, implemented a user-reporting mechanism called the "seems like AI slop" button to combat the issue. This initiative has seen over one million clicks, resulting in a 40% reduction in views for…

Take-Two Interactive, the parent company of Rockstar Games, is taking legal action against Microsoft and Discord following a major leak of in-game footage from the highly anticipated Grand Theft Auto VI, set for release in November. The company has filed…

Apple Music is set to introduce a 'Made With AI' label to help users identify tracks with substantial generative AI involvement, in response to a surge of AI-generated music flooding streaming platforms. This label will be mandatory for record labels and…

A researcher uncovered a hidden feature on the AliExpress website where audio fingerprinting scripts operate without user knowledge. This issue started when their Bluetooth headphones disconnected due to the AliExpress homepage loading. Upon investigation,…

OpenAI has released an update for its macOS desktop application that integrates ChatGPT with Apple Messages, allowing the AI to autonomously manage iMessage and SMS communications. This requires significant system access and high privileges, raising privacy…

TikTok and its parent company, ByteDance, have reached a $400 million settlement with the U.S. Department of Justice over allegations of violations of the Children's Online Privacy Protection Act (COPPA). The lawsuit, originating in 2024, accused TikTok of…

Iranian hackers successfully shut down a UK power plant for four days, marking an unprecedented event in cyberattacks against UK energy infrastructure, coinciding with similar attacks on water facilities across 12 US states. British officials have not…

The Security Affairs newsletter Round 591 discusses recent developments in cybersecurity and highlights significant articles concerning malware, hacking, and intelligence. Key issues include the upgrade of ToxicPanda malware affecting 16 countries, various…

Adversa AI has introduced a new attack method called Cryptographic Context Injection, which allows hackers to steal chat histories from AI systems like xAI's Grok and Google's Gemini without any user interaction. This technique involves sending encrypted…

A critical security vulnerability has been identified in the EverShop software, allowing for unauthenticated account takeovers, tracked as CVE-2026-72843. This flaw has a CVSS score of 9.3. The vulnerability affects versions prior to 2.2.1, allowing…

The content discusses a critical security alert regarding the Manic Android malware, which has been detected as capable of OS command injection through the Zimbra Collaboration Suite (ZCS). This malware targets users in Ukraine, Russia, and Europe and…

ToxicPanda 2.0, evolving from its previous iteration, targets 349 financial apps across 16 countries, vastly expanding from just 16 apps previously. The malware starts as a dropper, disguising itself to gain VPN permissions, thereby blocking Google Play…

TikTok has agreed to a $400 million settlement in a U.S. lawsuit concerning child privacy violations. The case was centered on allegations that TikTok failed to protect the personal information of children under 13, thus breaching the Children's Online…

Kaspersky researchers reported a new form of malware targeting Android-based car head units, found in June 2026. The malware exploits the car's official update system to install proxy software, linking affected vehicles to the BADBOX botnet. This marks the…