security researchers Archives

Winsage

June 5, 2026

Microsoft dropped Copilot+ PC branding for Windows 11’s most powerful AI laptop, and nobody complained

In May 2024, Microsoft launched Copilot+ PCs featuring local AI acceleration and Qualcomm's Snapdragon X Elite chips. However, during the introduction of the Surface Laptop Ultra, powered by NVIDIA’s RTX Spark platform, Microsoft did not mention the Copilot+ branding. The Surface Laptop Ultra is described as the most powerful AI-centric Windows laptop, yet it lacks the Copilot+ label, raising questions about the branding strategy. Initially, Copilot+ PCs were marketed as the fastest and most intelligent Windows PCs, requiring specific hardware specifications. The Recall feature, intended as a "photographic memory," faced privacy issues, leading to its retraction and redesign. In 2025, Microsoft integrated Copilot into various Windows applications, resulting in user backlash and a decline in the brand's perception. The Surface Laptop Ultra is confirmed as a Copilot+ PC internally, but Microsoft chose not to use the branding publicly, likely due to NVIDIA's branding interests. Additionally, inconsistencies in hardware requirements for Copilot+ PCs have created confusion among consumers. Microsoft may need to consider a rebranding of Windows as it focuses on local AI development and improving performance.

AppWizard

June 3, 2026

Weedhack malware campaign infects 116,000 mod-hungry Minecraft players systems through SEO poisoning and YouTube

Cybercriminals are using YouTube to distribute malware targeting Minecraft users, identified as Weedhack by McAfee Labs. This malware disables security defenses and allows attackers remote access to infected computers. The campaign offers both free and paid versions, making it accessible, especially to younger audiences. It also has the capability to steal Minecraft accounts, increasing its appeal.

Tech Optimizer

May 21, 2026

Nvidia tells users to update GPU drivers now or face possible attack — here’s what we know

NVIDIA has released an update to its GPU display drivers that addresses 14 vulnerabilities across its product lines, including GeForce, RTX, Quadro, Tesla, NVS, vGPU, and Cloud Gaming software. The most critical vulnerability is CVE‑2026‑24187, a high-severity use-after-free bug rated 8.8 out of 10, which could allow code execution, privilege escalation, data theft, or system crashes. Linux systems are vulnerable due to improper access to GPU resources at the kernel level, while Windows systems are at risk from a timing flaw. Two vulnerabilities in NVIDIA’s Unified Virtual Memory subsystem on Linux could lead to denial-of-service attacks without elevated permissions. The vGPU software also received patches for vulnerabilities in its virtual GPU manager component. Users can download the updated drivers from the NVIDIA Driver Downloads page or the NVIDIA Licensing Portal, with Windows users needing version 569.49 or newer and Linux users needing version 590.48.01. Users are advised to maintain their antivirus programs for enhanced security. NVIDIA thanked external security researchers for their responsible disclosure of these vulnerabilities.

Tech Optimizer

May 20, 2026

PoC Exploit Released for 20-Year Old PostgreSQL RCE Vulnerability

A proof-of-concept exploit has been developed for CVE-2026-2005, a remote code execution vulnerability in the pgcrypto extension of PostgreSQL. This vulnerability, stemming from legacy code, allows attackers to trigger a heap-based buffer overflow through specially crafted PGP messages, enabling arbitrary memory read and write operations. Successful exploitation can escalate privileges to PostgreSQL superuser status and execute commands on the operating system. The exploit targets PostgreSQL instances compiled from a specific vulnerable commit and circumvents protections like Address Space Layout Randomization (ASLR). It involves corrupting heap memory structures, leading to a controlled pointer leak that reveals the heap layout. Security researcher Varik Matevosyan has published the PoC on GitHub, demonstrating the exploitation process. The exploit requires a compatible PostgreSQL binary and utilizes Python-based tools for interaction. Organizations are advised to review their PostgreSQL deployments, disable unnecessary extensions, and apply security updates to mitigate risks.

Winsage

May 20, 2026

MSHTA abuse helps malware hide in Windows processes

Bitdefender's research highlights the use of Microsoft's MSHTA utility in malware attacks, noting its default activation in Windows systems. Cybercriminals exploit MSHTA to execute malicious scripts under the guise of legitimate processes, linking it to various malware families like LummaStealer and PurpleFox. The study reports a rise in MSHTA-related detections, indicating a shift towards "living-off-the-land" tactics that utilize legitimate tools to evade security alerts. Social engineering is identified as a common entry point for attacks, employing deceptive methods such as fake software downloads and phishing links. MSHTA can retrieve and execute additional payloads through multi-stage chains, complicating detection efforts. The attacks target sensitive information, including credentials and financial data, and the continued presence of MSHTA poses risks as it allows threat actors to conceal malicious actions. To mitigate these threats, organizations are advised to restrict or disable legacy scripting tools and exercise caution with untrusted downloads. The report emphasizes the challenge of detecting unusual behaviors associated with legitimate utilities in the context of cyber threats.

Winsage

May 14, 2026

Windows 11 and Microsoft Edge hacked at Pwn2Own Berlin 2026

On the inaugural day of Pwn2Own Berlin 2026, a total of ,000 was awarded to security researchers for exploiting 24 unique zero-day vulnerabilities. Orange Tsai earned ,000 for chaining four logic bugs to achieve a sandbox escape on Microsoft Edge. Windows 11 was targeted by Angelboy, TwinkleStar03, Marcin Wiązowski, and Kentaro Kawane, each earning ,000 for demonstrating new privilege escalation zero-days. Valentina Palmiotti earned ,000 for rooting Red Hat Linux for Workstations and an additional ,000 for a zero-day in the NVIDIA Container Toolkit. Other notable exploits included k3vg3n earning ,000 for taking down LiteLLM, Satoki Tsuji and haehae earning ,000 for exploiting NVIDIA Megatron Bridge zero-days, Compass Security and maitai earning ,000 each for hacking OpenAI's Codex, haehae earning ,000 for a Chroma zero-day, and STARLabs SG earning ,000 for exploiting a LM Studio zero-day. The DEVCORE Research Team leads the competition with ,000 in earnings, followed by Valentina Palmiotti with ,000. The contest is held at the OffensiveCon conference from May 14 to May 16, with over ,000,000 in cash and prizes available. Participants must target fully patched products and demonstrate arbitrary code execution. Vendors have a 90-day window to release security fixes after zero-day flaws are disclosed. Last year, the TrendMicro Zero Day Initiative awarded ,078,750 for 29 zero-day vulnerabilities.

Winsage

May 14, 2026

Dell confirms its SupportAssist software causes Windows BSOD crashes

Dell's SupportAssist software is causing blue-screen crashes on certain Windows systems, attributed to a recent update to the SupportAssist Remediation service, specifically version 5.5.16.0. Users experiencing these crashes are advised to uninstall or disable the service to resolve the issue. Dell has acknowledged the problem and is working on a solution. Uninstalling the service may result in the loss of system repair points created by Dell OS SupportAssist Recovery. Users still facing issues after uninstallation should contact Dell support. This incident follows previous software challenges faced by Dell, including blue screens from earlier SupportAssist versions and BIOS updates that prevented some laptops from booting. Additionally, vulnerabilities have been identified in the BIOSConnect feature of Dell SupportAssist, posing security risks.

Winsage

May 14, 2026

Microsoft MDASH finds Windows security flaws with AI | ETIH EdTech News

Microsoft has developed an AI security system called MDASH that has identified 16 vulnerabilities in Windows networking and authentication frameworks, with four classified as Critical, allowing for remote code execution. MDASH uses over 100 specialized AI agents to analyze code from multiple perspectives, improving the accuracy of vulnerability detection. The system has successfully detected all 21 planted vulnerabilities in a private testing environment without false positives, achieving a 96 percent recall rate against historical cases and an 88.45 percent success rate on the CyberGym benchmark. The identified vulnerabilities include issues in tcpip.sys and IKEEXT, which can be exploited remotely without needing credentials.

AppWizard

May 12, 2026

TrickMo Android Malware Targets Banking, Wallet, and Authenticator Apps

TrickMo, an Android banking malware, has re-emerged with enhanced stealth and operational capabilities, targeting banking, fintech, wallet, and authenticator apps. It retains its core device takeover abilities while improving stealth, persistence, and flexibility. The malware persuades users to grant accessibility permissions, allowing full remote control, including real-time screen viewing. A significant advancement is its shift to The Open Network (TON) for command-and-control communication, complicating takedown efforts. TrickMo has been actively deployed in France, Italy, and Austria since early 2026, using fake login overlays to capture credentials while recording user activity. It communicates through .adnl endpoints within the TON network and employs DNS-over-HTTPS for encrypted DNS queries. The malware's modular design includes a primary application as a loader and a dynamically loaded APK module called “dex.module” for malicious capabilities. It features network reconnaissance and tunneling capabilities, allowing commands like HTTP probing and establishing SSH tunnels, which can bypass fraud detection systems. Dormant features suggest potential for future capabilities without altering the core application. Indicators of compromise include specific SHA-256 hashes and package names associated with TrickMo's various components.

AppWizard

May 11, 2026

Google warns 7 million Android users to delete these 28 fake apps

28 Android applications were removed from the Google Play Store after being identified as scams by security researchers at ESET. These apps, part of a campaign called “CallPhantom,” falsely claimed to provide access to private call logs, SMS records, and WhatsApp activity. They attracted millions of downloads despite lacking legitimacy, offering fabricated data such as fake phone numbers and bogus call durations. Some apps charged users for “detailed reports” that either never arrived or contained nonsensical information. The apps did not steal phone data or install malware but instead promised illicit access and generated fictitious data. The primary targets of this scam were users in India and the Asia-Pacific region.