Tag: Cyber Threats

Aug 11, 2025Ravie LakshmananEncryption / Network Security

Cybersecurity researchers have discovered a fresh set of security issues in the Terrestrial Trunked Radio (TETRA) communications protocol, including in its proprietary end-to-end encryption (E2EE) mechanism that exposes the system to replay and brute-force attacks, and even decrypt encrypted traffic.

Details of the vulnerabilities – dubbed 2TETRA:2BURST – were presented at the Black Hat USA security conference last week by Midnight Blue researchers Carlo Meijer, Wouter Bokslag, and Jos Wetzels.

TETRA is a European mobile radio standard that’s widely used by law enforcement, military, transportation, utilities, and critical infrastructure operators. It was developed by the European Telecommunications Standards Institute (ETSI). It encompasses four encryption algorithms: TEA1, TEA2, TEA3, and TEA4.

The disclosure comes a little over two years after the Netherlands-based cybersecurity company discovered a set of security vulnerabilities in TETRA standard called TETRA:BURST, counting what was described as an “intentional backdoor” that could be exploited to leak sensitive information.

The newly discovered issues relate to a case of packet injection in TETRA, as well as an insufficient fix for CVE-2022-24401, one of the five TETRA:BURST issues, to prevent keystream recovery attacks. The identified issues are listed below –

• CVE-2025-52940 – TETRA end-to-end encrypted voice streams are vulnerable to replay attack. Furthermore, an attacker with no knowledge of the key may inject arbitrary voice streams, that are played back indistinguishably from authentic traffic by legitimate call recipients.
• CVE-2025-52941 – TETRA end-to-end encryption algorithm ID 135 refers to an intentionally weakened AES-128 implementation which has its effective traffic key entropy reduced from 128 to 56 bits, rendering it vulnerable to brute-force attacks.
• CVE-2025-52942 – End-to-end encrypted TETRA SDS messages feature no replay protection, allowing for arbitrary replay of messages towards either humans or machines.
• CVE-2025-52943 – TETRA networks that support multiple Air Interface Encryption algorithms are vulnerable to key recovery attacks since the SCK/CCK network key is identical for all supported algorithms. When TEA1 is supported, an easily recovered TEA1 key (CVE-2022-24402) can be used to decrypt or inject TEA2 or TEA3 traffic on the network.
• CVE-2025-52944 – The TETRA protocol lacks message authentication and therefore allows for the injection of arbitrary messages such as voice and data.
• ETSI’s fix for CVE-2022-24401 is ineffective in the prevention of keystream recovery attacks (No CVE, assigned a placeholder identifier MBPH-2025-001)

In a statement shared with WIRED, ETSI said the E2EE mechanism used in TETRA-based radios is not part of the ETSI standard, adding it was produced by The Critical Communications Association’s (TCCA) security and fraud prevention group (SFPG). ETSI also noted that purchasers of TETRA-based radios are free to deploy other solutions for E2EE on their radios.

The findings also coincide with the discovery of three flaws in the Sepura SC20 series of mobile TETRA radios that allow attackers with physical access to the device to achieve unauthorized code execution –

• CVE-2025-52945 – Defective file management restrictions
• CVE-2025-8458 – Insufficient key entropy for SD card encryption
• Exfiltration of all TETRA and TETRA E2EE key materials with the exception of the device-specific key K (no CVE, assigned a placeholder identifier MBPH-2025-003)

Patches for CVE-2025-52945 and CVE-2025-8458 are expected to be made available in the third quarter of 2025, necessitating that users are advised to implement enhanced TETRA key management policies. MBPH-2025-003, on the other hand, cannot be remediated due to architectural limitations.

“The vulnerabilities enable an attacker to gain code execution on a Sepura Gen 3 device,” the company said. “Attack scenarios featuring CVE-2025-8458 involve persistent code execution through access to a device’s SD card. Abuse of CVE-2025-52945 is even more straightforward as it requires only brief access to the device’s PEI connector.”

“From the premise of code execution, multiple attack scenarios are viable, such as exfiltration of TETRA key materials (MBPH-2025-003) or the implantation of a persistent backdoor into the radio firmware. This leads to the loss of confidentiality and integrity of TETRA communications.”

Aug 11, 2025Ravie LakshmananVulnerability / Network Security

Malicious actors have been observed exploiting a now-patched critical security flaw impacting Erlang/Open Telecom Platform (OTP) SSH as early as beginning of May 2025, with about 70% of detections originating from firewalls protecting operational technology (OT) networks.

The vulnerability in question is CVE-2025-32433 (CVSS score: 10.0), a missing authentication issue that could be abused by an attacker with network access to an Erlang/OTP SSH server to execute arbitrary code. It was patched in April 2025 with versions OTP-27.3.3, OTP-26.2.5.11, and OTP-25.3.2.20.

Then in June 2025, the U.S. Cybersecurity and Infrastructure Security Agency (CISA) added the flaw to its Known Exploited Vulnerabilities (KEV) catalog, based on evidence of active exploitation.

“At the heart of Erlang/OTP’s secure communication capabilities lies its native SSH implementation — responsible for encrypted connections, file transfers and most importantly, command execution,” Palo Alto Networks Unit 42 researchers Adam Robbie, Yiheng An, Malav Vyas, Cecilia Hu, Matthew Tennis, and Zhanhao Chen said.

“A flaw in this implementation would allow an attacker with network access to execute arbitrary code on vulnerable systems without requiring credentials, presenting a direct and severe risk to exposed assets.”

The cybersecurity company’s analysis of telemetry data has revealed that over 85% of exploit attempts have primarily singled out healthcare, agriculture, media and entertainment, and high technology sectors in the U.S., Canada, Brazil, India, and Australia, among others.

In the attacks observed, the successful exploitation of CVE-2025-32433 is followed by the threat actors using reverse shells to gain unauthorized remote access to target networks. It’s currently not known who is behind the efforts.

“This widespread exposure on industrial-specific ports indicates a significant global attack surface across OT networks,” Unit 42 said. “Analysis of affected industries demonstrates variance in the attacks.”

“Attackers are attempting to exploit the vulnerability in short, high-intensity bursts. These are disproportionately targeting OT networks and attempting to access exposed services over both IT and industrial ports.”

Aug 11, 2025Ravie Lakshmanan

This week, cyber attackers are moving quickly, and businesses need to stay alert. They’re finding new weaknesses in popular software and coming up with clever ways to get around security. Even one unpatched flaw could let attackers in, leading to data theft or even taking control of your systems. The clock is ticking—if defenses aren’t updated regularly, it could lead to serious damage. The message is clear: don’t wait for an attack to happen. Take action now to protect your business.

Here’s a look at some of the biggest stories in cybersecurity this week: from new flaws in WinRAR and NVIDIA Triton to advanced attack techniques you should know about. Let’s get into the details.

⚡ Threat of the Week

Trend Micro Warns of Actively Exploited 0-Day — Trend Micro has released temporary mitigations to address critical security flaws in on-premise versions of Apex One Management Console that it said have been exploited in the wild. The vulnerabilities (CVE-2025-54948 and CVE-2025-54987), both rated 9.4 on the CVSS scoring system, have been described as management console command injection and remote code execution flaws. There are currently no details on how the issues are being exploited in real-world attacks. Trend Micro said it “observed at least one instance of an attempt to actively exploit one of these vulnerabilities in the wild.”

• WinRAR 0-Day Under Active Exploitation — The maintainers of the WinRAR file archiving utility have released an update to address an actively exploited zero-day vulnerability. Tracked as CVE-2025-8088 (CVSS score: 8.8), the issue has been described as a case of path traversal affecting the Windows version of the tool that could be exploited to obtain arbitrary code execution by crafting malicious archive files. Russian cybersecurity vendor BI.ZONE, in a report published last week, said there are indications that the hacking group tracked as Paper Werewolf (aka GOFFEE) may have leveraged CVE-2025-8088 alongside CVE-2025-6218, a directory traversal bug in the Windows version of WinRAR that was patched in June 2025.
• New Windows EPM Poisoning Exploit Chain Detailed — New findings presented at the DEF CON 33 security conference showed that a now-patched security issue in Microsoft’s Windows Remote Procedure Call (RPC) communication protocol (CVE-2025-49760, CVSS score: 3.5) could be abused by an attacker to conduct spoofing attacks and impersonate a known server. The vulnerability essentially makes it possible to manipulate a core component of the RPC protocol and stage what’s called an EPM poisoning attack that allows unprivileged users to pose as a legitimate, built-in service with the goal of coercing a protected process to authenticate against an arbitrary server of an attacker’s choosing.
• BadCam Attack Targets Linux Webcams From Lenovo — Linux-based webcams from Lenovo, Lenovo 510 FHD and Lenovo Performance FHD, which are powered by a System on a Chip (SoC) and firmware made by the Chinese company SigmaStar, can be weaponized and turned into BadUSB vectors, allowing attackers to tamper with the firmware of the devices to execute malicious commands when connected to a computer. “This allows remote attackers to inject keystrokes covertly and launch attacks independent of the host operating system,” Eclypsium researchers Paul Asadoorian, Mickey Shkatov, and Jesse Michael said.
• The Far-Reaching Scale of VexTrio Revealed — A new analysis of VexTrio has unmasked it as a “cybercriminal organization with tendrils that are far-reaching,” operating dozens of businesses and front companies across Europe, while posing as a legitimate ad tech firm to conduct various types of fraud. The cyber fraud network is assessed to be active in its present form since at least 2017. That said, suspected key figures behind the scheme have been linked to scam reports and sketchy domains since 2004. VexTrio’s nerve center is Lugano, melding scam operations and traffic distribution schemes to maximize illicit revenue. It’s also the result of two businesses, Tekka Group and AdsPro Group, joining forces in 2020. “The merger created a formidable suite of commercial entities that touch every part of the ad tech industry,” Infoblox said. VexTrio is known for using traffic distribution systems (TDSes) to filter and redirect web traffic based on specific criteria, as well as relying on sophisticated DNS manipulation techniques like fast-fluxing, DNS tunneling, and domain generation algorithms (DGAs) to rapidly change the IP addresses associated with their domains, establish covert command-and-control (C2) communication, and maintain persistent access with infected systems. Campaigns orchestrated the threat actor to leverage TDSes to hijack web users from compromised websites and redirect them to a variety of malicious destinations, from tech support scams and fake updates to phishing domains and exploit kits. The use of commercial entities to run the traffic distribution schemes offers several advantages to threat actors, both from an operational perspective as well as avoiding scrutiny from the infosec community and law enforcement by maintaining a veneer of legitimacy. The system works like any other ad tech network, only it’s malicious in nature. The threat actors pay VexTrio-controlled firms as if they were legitimate customers, receiving a steady supply of hijacked traffic and unsuspecting victims through TDSes for a variety of threats, from cryptocurrency scams and fake captcha schemes. “VexTrio employs a few hundred people globally. It’s unclear how much the average VexTrio employee knows about the true business model,” Infoblox said. The arrangement has proven to be extremely lucrative for VexTrio operators, who have been found leading a lavish lifestyle, sharing on social media about expensive cars and other luxuries.
• Multiple Flaws Patched in NVIDIA Triton Patched — Nvidia has patched a trio of vulnerabilities in its Triton inference server that could give unauthenticated remote attackers a way to take full control of susceptible servers. The new Triton vulnerabilities underscore a broader and rapidly growing category of AI-related threats that organizations must now factor into their security postures. With AI and ML tools becoming deeply embedded in critical business workflows, the attack surface has expanded in ways that traditional security frameworks aren’t always equipped to handle. The emergence of new threats like AI supply chain integrity, model poisoning, prompt injection, and data leakage signals the need for securing the underlying infrastructure and practicing defense-in-depth.

‎️‍🔥 Trending CVEs

Hackers are quick to jump on newly discovered software flaws – sometimes within hours. Whether it’s a missed update or a hidden bug, even one unpatched CVE can open the door to serious damage. Below are this week’s high-risk vulnerabilities making waves. Review the list, patch fast, and stay a step ahead.

This week’s list includes — CVE-2025-8088 (WinRAR), CVE-2025-55188 (7-Zip), CVE-2025-4371 (Lenovo 510 FHD and Performance FHD web cameras), CVE-2025-25050, CVE-2025-25215, CVE-2025-24922, CVE-2025-24311, CVE-2025-24919 (Dell ControlVault3), CVE-2025-49827, CVE-2025-49831 (CyberArk Secrets Manager), CVE-2025-6000 (HashiCorp Vault), CVE-2025-53786 (Microsoft Exchange Server), CVE-2025-30023 (Axis Communications), CVE-2025-54948, CVE-2025-54987 (Trend Micro Apex One Management Console), CVE-2025-23310, CVE-2025-23311, CVE-2025-23319 (NVIDIA Triton), CVE-2025-54574 (Squid Web Proxy), CVE-2025-7025, CVE-2025-7032, and CVE-2025-7033 (Rockwell Automation Arena Simulation), CVE-2025-54253, CVE-2025-54254 (Adobe Experience Manager Forms), CVE-2025-24285 (Ubiquiti UniFi Connect EV Station), CVE-2025-38236 (Linux Kernel), CVE-2025-2771, CVE-2025-2773 (BEC Technologies routers), CVE-2025-25214, CVE-2025-48732 (WWBN AVideo), CVE-2025-26469, and CVE-2025-27724 (MedDream PACS Premium).

📰 Around the Cyber World

• NVIDIA Rejects Backdoor Claims — GPU maker NVIDIA has rejected accusations that it has built backdoors or kill switches in its chips. “There are no back doors in NVIDIA chips. No kill switches. No spyware. That’s not how trustworthy systems are built—and never will be,” Nvidia Chief Security Officer David Reber Jr. said. The development came after the Cyberspace Administration of China (CAC) said it held a meeting with NVIDIA over “serious security issues” in the company’s chips and claimed that U.S. artificial intelligence (AI) experts “revealed that NVIDIA’s computing chips have location tracking and can remotely shut down the technology.” A kill switch in a chip would be “a permanent flaw beyond user control, and an open invitation for disaster,” Reber Jr. added.
• Attackers Compromise Target Within 5 Minutes — Threat actors successfully compromised corporate systems within just five minutes using a combination of social engineering tactics and rapid PowerShell execution. The incident demonstrates how cybercriminals are weaponizing trusted business applications to bypass traditional security measures. “The Threat Actor targeted around twenty users, impersonating IT support personnel, and successfully convinced two users to grant remote access to their system using the Windows native Quick Assist remote support tool,” NCC Group said. “In less than five minutes, the Threat Actor executed PowerShell commands that led to the download of offensive tooling, malware execution and the creation of persistence mechanisms.” The attack was detected and stopped before it could have led to a bigger infection.
• Companies Drowning in Threat Intel — A new study commissioned by Google Cloud found that an “overwhelming volume of threats and data combined with the shortage of skilled threat analysts” are making companies more vulnerable to cyber attacks and keeping them stuck in a reactive state. “Rather than aiding efficiency, myriad [threat intelligence] feeds inundate security teams with data, making it hard to extract useful insights or prioritize and respond to threats. Security teams need visibility into relevant threats, AI-powered correlation at scale, and skilled defenders to use actionable insights, enabling a shift from a reactive to a proactive security posture,” the study found. The survey was conducted with 1,541 senior IT and cybersecurity leaders at enterprise organizations in North America, Europe, and Asia Pacific.

• New EDR Killer Spotted — Malware capable of terminating antivirus software and obfuscated using commercial packers like HeartCrypt are being used in ransomware attacks involving BlackSuit, RansomHub, Medusa, Qilin, DragonForce, Crytox, Lynx, and INC. Posing as a legitimate utility, the EDR killer looks for a driver with a five-letter random name that’s signed with a compromised certificate to achieve its goals. If found, the malicious driver is loaded into the kernel, as required to perform a bring your own vulnerable driver (BYOVD) attack and achieve kernel privileges required to turn off security products. The exact list of antivirus software to be terminated varies among samples. It’s believed to be an evolution of EDRKillShifter, developed by RansomHub. “Multiple new variants of a malicious driver that first surfaced in 2022 are circulating in the wild,” Symantec warned earlier this January. “The driver is used by attackers to attempt to disable security solutions.” The fact that multiple ransomware actors are relying on variants of the same EDR killer tool alludes to the possibility of a common seller or some sort of an “information/tool leakage between them.”
• Ransomware Continues to Evolve — Threat intel firm Analyst1 has published a profile of Yaroslav Vasinskyi, a Ukrainian national and member of the REvil gang that broke into Kaseya in 2021. Meanwhile, the ransomware landscape continues to be volatile as ever, replete with rebrands and abrupt cessation of activities amid continued law enforcement takedowns: BlackNevas (aka Trial Recovery) is assessed to be a derivative of Trigona, while one affiliate named “hastalamuerte” alleged that the Qilin group had conducted an exit scam, defrauding them of $48,000. Another user, operating under the handle “Nova,” publicly leaked the Qilin affiliate panel, including login credentials, further exposing the group’s operational security weaknesses. RansomHub, Babuk-Bjorka, FunkSec, BianLian, 8Base, CACTUS, Hunters International, and LockBit are among the groups that have stopped publishing new victims, indicating an increasingly fragmented ransomware ecosystem. “The rapid succession of events following the disappearance of RansomHub and the subsequent rise – and apparent turbulence – within Qilin’s operations underscore the dynamic volatility of today’s ransomware ecosystem,” Dark Atlas said. “The internal chaos and alleged exit scam within Qilin […] reveal deep fissures in trust and operational security among ransomware collectives, further compounded by active interference from law enforcement and rival groups.”
• Turkish Organizations Targeted by SoupDealer — Banks, ISPs, and mid-level organizations in Türkiye are being targeted by phishing campaigns that deliver a new Java-based loader called SoupDealer. “When this malware is executed, it uses advanced persistence mechanisms – including downloading TOR to establish communication with the C2 panel and scheduling tasks for automatic execution – to ensure the device is located in Türkiye and being used in Turkish,” Malwation said. “It then sends various information based on signals from the command-and-control server and gains full control over the device.”
• Spark RAT Detailed — Cybersecurity researchers have detailed the inner workings of an open-source RAT called Spark RAT that’s capable of targeting Windows, Linux, and macOS systems. It allows an attacker to remotely commandeer a compromised endpoint by establishing communications with C2 infrastructure and awaiting further instructions from an operator. “All the desirable RAT features are present, with the perhaps notable absence of Remote Desktop-like functionality,” F5 Labs said. “These factors have combined to make SparkRAT an attractive offensive tool choice, as is evidenced by the documented instances of its use in threat campaigns.”
• Threat Actors’ Use of SVG Files Increase — Cybercriminals are turning Scalable Vector Graphics (SVG) files into potent weapons by embedding malicious JavaScript payloads that can bypass traditional security measures. Phishing attacks adopting the technique have revolved around convincing targets to open an SVG file, triggering the execution of the JavaScript code in the web browser, which then redirects them to a phishing site designed to steal credentials. “Instead of storing pixel data, SVGs use XML-based code to define vector paths, shapes, and text,” Seqrite said. “This makes them ideal for responsive design, as they scale without losing quality. However, this same structure allows SVGs to contain embedded JavaScript, which can execute when the file is opened in a browser – something that happens by default on many Windows systems.” SVG image files are also being used as a malware delivery vector in campaigns where adult sites have been found seeding obscured SVG payloads that leverage JSFuck to covertly endorse Facebook posts promoting the sites, ThreatDown found.
• Scams Targeting Elderly Led to $700 million Losses in 2024 — Americans aged 60 and older lost a staggering $700 million to online scams in 2024, signaling a steep rise in fraud targeting older adults. “Most notably, combined losses reported by older adults who lost more than $100,000 increased eight-fold, from $55 million in 2020 to $445 million in 2024,” the U.S. Federal Trade Commission (FTC) said. “While younger consumers also have reported these scams, older adults were much more likely to report these extraordinarily high losses.” The development came as authorities from the Philippines detained 20 Chinese nationals who were operating a crypto scam center in Pasay City. Thai police have also apprehended 18 Chinese nationals who were operating a scam call center in the city of Chiang Mai that targeted other Chinese speakers and operated for three months from a rented house.

• Embargo Ransomware Made About $34.2 million — Embargo ransomware is associated with about $34.2 million in cryptocurrency transactions since popping up around April 2024, with the majority of the victims located in the United States in the healthcare, business services, and manufacturing sectors. Unlike other traditional ransomware-as-a-service (RaaS) groups, Embargo retains control over infrastructure and payment negotiations and tends to avoid tactics like triple extortion and victim harassment that draw attention to itself. The attacks involve using phishing emails and drive-by downloads delivered via malicious websites as initial access vectors to disable security tools, turn off recovery options, and encrypt files. “Embargo may be a rebranded or successor operation to BlackCat (ALPHV) based on multiple technical and behavioral similarities – including using the Rust programming language, a similarly designed data leak site, and on-chain overlaps via shared wallet infrastructure,” TRM Labs said. “Embargo launders ransom proceeds through intermediary wallets, high-risk exchanges, and sanctioned platforms such as Cryptex.net. Approximately $18.8 million remains dormant in unattributed wallets — a pattern that likely reflects deliberate evasion tactics.” The links to BlackCat stem from on-chain overlaps, with historical BlackCat-linked addresses funneling funds to wallet clusters associated with Embargo victims. Technical similarities include the use of the Rust programming language, similar encryption toolkits, and the design of their data leak sites.
• Microsoft to Block File Access via FPRPC — Microsoft has announced that the Microsoft 365 apps for Windows will start blocking access to files via the insecure FPRPC legacy authentication protocol by default starting late August. “Microsoft 365 apps will block insecure file open protocols like FPRPC by default starting version 2508, with new Trust Center settings to manage these protocols,” the company said. “These changes enhance security by reducing exposure to outdated technologies like FrontPage Remote Procedure Call (FPRPC), FTP, and HTTP.” Separately, Microsoft has also announced that it intends to retire support for inline SVG images in Outlook for Web and new Outlook for Windows starting September 2025. “This change enhances security and aligns with current email client behavior, which already restricts inline SVG rendering,” the company said.
• Nearly 30K Exchange Server Instances Vulnerable to CVE-2025-53786 — A little over 29,000 Microsoft Exchange email servers are missing an April 2025 hotfix for a recently disclosed security vulnerability (CVE-2025-53786) that allows attackers to escalate access from on-prem servers to online cloud environments. As of August 10, 2025, the countries with the most exposures are the U.S., Germany, Russia, France, the U.K., and Austria, per the Shadowserver Foundation.
• ScarCruft Linked to Ransomware Attack for the First Time — The North Korean threat actor known as ScarCruft (aka APT37), which has a history of deploying RokRAT, has been linked to an attack chain that has leveraged a malicious LNK file embedded in a RAR archive to deliver a stealer (LightPeek and FadeStealer), backdoor (NubSpy and CHILLYCHINO), and ransomware (VCD Ransomware). “It further underscores the group’s persistent reliance on real-time messaging infrastructure, exemplified by NubSpy’s use of PubNub as its command-and-control (C2) channel,” S2W said. The attack has been attributed to ChinopuNK, a sub-cluster within ScarCruft known for deploying the Chinotto malware. The activity is a “notable deviation” from the group’s historical focus on espionage. “This suggests a potential shift toward financially motivated operations, or an expansion of operational goals that now include disruptive or extortion-driven tactics,” the company added.
• EDR-on-EDR Violence to Disable EDR Software — Cybersecurity researchers have uncovered a troubling new attack vector where threat actors are weaponizing free trials of endpoint detection and response (EDR) software to disable existing security tools – a phenomenon dubbed EDR-on-EDR violence, or bring your own EDR aka BYOEDR. “It turns out that one of the ways to disable EDR is with a free trial of EDR,” researchers Ezra Woods and Mike Manrod said. “This is accomplished by removing exclusions and then adding the hash of the existing AV/EDR as a blocked application.” Making matters worse, the research found that it’s possible to abuse the RMM-like features of EDR products to facilitate command shell access.
• 2 Founder of Samourai Wallet Plead Guilty to Money Laundering — Two senior executives and founders of the Samourai Wallet cryptocurrency mixer have pleaded guilty to charges involving washing more than $200 million worth of crypto assets from criminal proceeds and concealing the nature of illicit transactions using services like Whirlpool and Ricochet. Samourai CEO Keonne Rodriguez and CTO William Lonergan Hill were arrested last year after the U.S. Federal Bureau of Investigation (FBI) took down their service. As part of their plea agreements, Rodriguez and Hill have also agreed to forfeit $237,832,360.55. “The defendants created and operated a cryptocurrency mixing service that they knew enabled criminals to wash millions in dirty money, including proceeds from cryptocurrency thefts, drug trafficking operations, and fraud schemes,” the U.S. Department of Justice (DoJ) said. “They did not just facilitate this illicit movement of money, but also encouraged it.”
• Tornado Cash Founder Convicted of Operating a Money Transmitting Business — Roman Storm, a co-founder of the cryptocurrency mixing service Tornado Cash, was found guilty of conspiring to operate an unlicensed money-transmitting business. However, the jury failed to reach a ruling on the more significant charges of conspiracy to commit money laundering and to violate sanctions. “Roman Storm and Tornado Cash provided a service for North Korean hackers and other criminals to move and hide more than $1 billion of dirty money,” the DoJ said. Storm is set to be sentenced later this year and faces a maximum prison sentence of five years. The development came as the U.S. Treasury Department dropped its appeal against a court ruling that forced it to lift sanctions against Tornado Cash last month. Tornado Cash was delisted from the Specially Designated National and Blocked Persons (SDN) list earlier this March. The service was sanctioned in 2022 for its alleged links to cybercriminals and for having “repeatedly failed to impose effective controls” to prevent money laundering.
• Microsoft SharePoint Flaws Exploited to Drop China Chopper and ANTSWORD — Microsoft revealed that Chinese state-sponsored hackers had exploited new vulnerabilities in SharePoint to breach the computer systems of hundreds of companies and government agencies, including the National Nuclear Security Administration and the Department of Homeland Security. According to ProPublica, support for SharePoint is handled by a China-based engineering team that has been responsible for maintaining the software for years. Microsoft said the China-based team “is supervised by a US-based engineer and subject to all security requirements and manager code review. Work is already underway to shift this work to another location.” It’s unclear if Microsoft’s China-based staff had any role in the SharePoint hack. Attacks exploiting the SharePoint flaws (CVE-2025-49706 and CVE-2025-53770) have been observed performing unauthenticated code execution, extracting cryptographic keys, and deploying web shells like China Chopper and ANTSWORD. “The use of AntSword and China Chopper in the mid-2025 SharePoint exploitation campaigns aligns with tooling observed in prior incidents,” Trustwave said. “Notably, in 2022, the same ANTSWORD and China Chopper were also observed to be deployed in an incident related to ProxyNotShell RCE vulnerabilities.
• E.U. Law Protecting Journalists from Spyware Goes into Effect — A new law in the European Union, called the European Media Freedom Act (EMFA), has taken effect starting August 8, 2025, seeking to promote independence, safeguard media against unjustified online content removal by very large online platforms, and protect journalistic sources, including against the use of spyware. However, the European Centre for Press and Media Freedom (ECPMF) said it’s “deeply concerned that many national governments are neither prepared nor politically willing to make the required legislative changes,” adding “this lack of commitment poses a serious risk to the EMFA’s effectiveness.”
• Israel Created Azure-Backed System to Store Palestinian Communications — Israel’s elite military surveillance agency, Unit 8200, stored vast volumes of intercepted Palestinian phone calls on Microsoft’s Azure cloud servers, according to a joint investigation by The Guardian, +972 Magazine, and Local Call. The massive phone surveillance operation intercepted and tracked all phone calls and messages sent across Palestine and was hosted in a segregated part of Azure. The cloud-based system is believed to have become operational in 2022. “Thanks to the control it exerts over Palestinian telecommunications infrastructure, Israel has long intercepted phone calls in the occupied territories,” The Guardian reported. “But the indiscriminate new system allows intelligence officers to play back the content of cellular calls made by Palestinians, capturing the conversations of a much larger pool of ordinary civilians.”
• South Korea Targeted by Makop Ransomware — Users in South Korea have been targeted by Makop ransomware attacks that leverage remote desktop protocol (RDP) as an entry point, shifting from its previous distribution strategy of relying on fake resumes or emails related to copyrights. “It is worth noting that the use of RDP in the initial access phase and the installation of various tools from NirSoft and Mimikatz with an installation path of ‘mimik’ are the same as what the Crysis ransomware threat actor did when installing the Venus ransomware,” AhnLab said. “This suggests the possibility that the same threat actor is behind the Crysis, Venus, and recent Makop ransomware attacks.”
• WhatsApp Rolls Out New Feature to Tackle Scams — WhatsApp is introducing a new security feature that will help users spot potential scams when they are being added to a group chat by someone who is not in their contact list by serving additional information and options to exit the group. The messaging platform said it’s also exploring ways to caution people when they are individually contacted by people not in their contacts. This includes showing more context about who has messaged, so users can make an informed decision. The Meta-owned company said it also took down over 6.8 million WhatsApp accounts linked to criminal scam centers based in Southeast Asia targeting people across the internet and around the world. “These scam centers typically run many scam campaigns at once – from cryptocurrency investments to pyramid schemes,” the company said. “The scammers used ChatGPT to generate the initial text message containing a link to a WhatsApp chat, and then quickly directed the target to Telegram, where they were assigned a task of liking videos on TikTok. The scammers attempted to build trust in their scheme by sharing how much the target has already ‘earned’ in theory, before asking them to deposit money into a crypto account as the next task.”
• Praetorian Releases ChromeAlone — Cybersecurity company Praetorian has released a tool called ChromeAlone that transforms Chromium browsers into a C2 framework and can be implanted and used in place of conventional tools like Cobalt Strike. The program offers the ability to steal browser credentials and session cookies, launch executables on the host from Chrome, phish for WebAuthn requests for physical security tokens like YubiKeys or Titan Security Keys, and offer EDR resistance. Separately, Praetorian also found that it’s possible to abuse Traversal Using Relays around NAT (TURN) servers used by conferencing apps like Zoom and Microsoft Teams as a new C2 evasion method called ‘Ghost Calls’ to tunnel traffic through trusted infrastructure. This is accomplished by means of a tool called TURNt. “This approach allows operators to blend interactive C2 sessions into normal enterprise traffic patterns, appearing as nothing more than a temporarily joined online meeting,” Praetorian noted, stating the approach uses legitimate credentials, WebRTC, and custom tooling to get around existing defenses.
• New Jailbreak Against AI Chatbots Employs Information Overload — AI chatbots like OpenAI ChatGPT and Google Gemini can be derived into generating illicit instructions for making a bomb or hacking an ATM if the prompt is made complicated, full of academic jargon, and cites non-existent sources. That’s according to a new paper authored by a team of researchers from Intel, Boise State University, and the University of Illinois at Urbana-Champaign. The LLM jailbreaking technique called InfoFlood “transforms malicious queries into complex, information-overloaded queries capable of bypassing built-in safety mechanisms,” the paper explained. “Specifically, InfoFlood: (1) uses linguistic transformations to rephrase malicious queries, (2) identifies the root cause of failure when an attempt is unsuccessful, and (3) refines the prompt’s linguistic structure to address the failure while preserving its malicious intent.”
• Israeli spyware vendor Candiru is still active — Cybersecurity firm Recorded Future has discovered new infrastructure for managing and delivering Candiru’s DevilsTongue spyware. “Eight distinct clusters were identified, with five being likely still active, including those linked to Hungary and Saudi Arabia,” it said. “One cluster tied to Indonesia was active until November 2024, and two associated with Azerbaijan have uncertain status due to a lack of identified victim-facing infrastructure.”

🎥 Cybersecurity Webinars

• AI Threats Are Real—Learn How to Secure Every Agent Now: AI-powered shadow agents are becoming a serious security threat. Deployed without oversight, these invisible entities have access to sensitive data, making them prime targets for attackers. In this session, we’ll explore how these agents emerge, why they’re risky, and how to take control before they cause harm.
• How AI-Fueled Attacks are Targeting Identity—Learn to Stop Them: AI is changing the way cyberattacks happen, making traditional defenses obsolete. In this webinar, Karl Henrik Smith from Okta explains how AI is targeting identity security and how you can protect your organization from these new threats. Learn how to adapt your defenses for the AI-driven future.
• What You’re Missing in Python Security: 2025’s Must-Know Threats: In 2025, securing your Python supply chain is more critical than ever. With increasing threats like repojacking, typosquatting, and known vulnerabilities in core Python infrastructure, simply relying on “pip install and pray” won’t cut it. Join our webinar to learn how to protect your Python projects, tackle current supply chain risks, and explore practical solutions to safeguard your code with industry-leading tools like Sigstore and Chainguard. Take action now to secure your Python environment and stay ahead of emerging threats.

🔧 Cybersecurity Tools

• DoomArena is a modular, plug-in framework for testing AI agents against evolving security threats. It works with platforms like τ-Bench, BrowserGym, and OSWorld, allowing realistic simulations of attacks such as prompt injections or malicious data sources. Its design separates attack logic from environments, making tests reusable across tasks, and supports detailed threat models, multiple attack types, and custom success checks to help identify vulnerabilities and evaluate defenses.
• Yamato Security, a volunteer-led group in Japan, has released a suite of open-source tools aimed at strengthening digital forensics and threat hunting. The lineup includes Hayabusa for Sigma-based Windows log analysis, Takajo for parsing Hayabusa results, Suzaku for cloud log forensics, and WELA for auditing Windows Event Logs, supported by detailed configuration guides. Also in the toolkit is SigmaOptimizer-UI, a user-friendly interface that streamlines the creation, testing, and refinement of Sigma rules from real-world logs, incorporating automated checks and optional LLM-powered enhancements.

Disclaimer: These newly released tools are for educational use only and haven’t been fully audited. Use at your own risk—review the code, test safely, and apply proper safeguards.

🔒 Tip of the Week

Boost Your Threat Detection with Easy, Free Tools — Cybersecurity isn’t just about defending against attacks—it’s also about detecting them early. One of the most effective ways to stay ahead of threats is by setting up real-time monitoring. Free tools like UptimeRobot allow you to monitor your website or systems for unexpected downtime, a common sign of an attack. By receiving instant alerts, you can act quickly if something goes wrong.

Another simple yet powerful step is running regular vulnerability scans. Qualys Community Edition is a free tool that helps you identify weak spots in your network or website. Regular scans will help you spot problems before attackers can exploit them, keeping your defenses strong.

Endpoint protection is equally important. While Windows Defender provides solid security, you can take it a step further with OSSEC, an open-source intrusion detection system. OSSEC monitors your devices for unusual behavior, helping catch threats that traditional antivirus software might miss.

Lastly, staying aware of malicious actors is key. Use resources like AlienVault Open Threat Exchange (OTX) to track known harmful IP addresses and domains. These free databases keep you informed about the latest threats targeting your network, allowing you to block risky traffic before it causes harm.

By integrating these free tools into your routine, you’ll significantly enhance your ability to detect and respond to cyber threats quickly and effectively.

Conclusion

As we wrap up this week’s cybersecurity update, remember that staying informed is your best defense. The threats are real, and the stakes are high—but with the right steps, your organization can stay ahead of attackers. Regular updates, timely patches, and continuous monitoring are your first line of defense. Keep working to build a culture of security, and always be ready to adapt to the changing landscape.

We’ll be back next week with more insights, so keep those systems secure and stay vigilant. Until then, stay proactive, stay safe, and don’t let your guard down. Cyber threats wait for no one.

The Evolution of Exposure Management

Most security teams have a good sense of what’s critical in their environment. What’s harder to pin down is what’s business-critical. These are the assets that support the processes the business can’t function without. They’re not always the loudest or most exposed. They’re the ones tied to revenue, operations, and delivery. If one goes down, it’s more than a security issue – It’s a business problem.

Over the past year since publishing our 4-step approach to mapping and securing business-critical assets, my team and I have had the opportunity to engage deeply with dozens of customer workshops across multiple industry verticals, including finance, manufacturing, energy, and more. These sessions have revealed valuable insights into how organizations are evolving their security posture.

This article takes an updated look at that approach, incorporating what we have learned along the way, helping organizations align exposure management strategy with business priorities. What began as a theoretical 4-step approach has matured into a proven methodology with measurable results. Organizations implementing this framework have reported remarkable efficiency gains—some reducing remediation efforts by up to 96% while simultaneously strengthening their security posture where it matters most.

Our engagement with CISOs, security directors, and increasingly, CFOs and business executives, has revealed consistent patterns across industries. Security teams struggle not with identifying vulnerabilities but with determining which ones pose genuine business risk. Meanwhile, business leaders want assurance that security investments protect what matters most—but often lack a framework to communicate these priorities effectively to technical teams.

The methodology we’ve refined bridges this gap, creating a common language between security practitioners and business stakeholders. The lessons that follow distill what we’ve learned through implementing this approach across diverse organizational contexts. They represent not just theoretical best practices, but practical insights gained through successful real-world applications.

Lesson 1: Not All Assets Are Created Equal

What We Discovered: Most security teams can identify what’s technically critical, but struggle to determine what’s business-critical. The difference is significant – business-critical assets directly support revenue generation, operations, and service delivery.

Key Takeaway: Focus your security resources on systems that, if compromised, would create actual business disruption rather than just technical issues. Organizations that implemented this targeted approach reduced remediation efforts by up to 96%.

Lesson 2: Business Context Changes Everything

What We Discovered: Security teams are drowning in signals – vulnerability scans, CVSS scores, and alerts from across the technology stack. Without business context, these signals lack meaning. A “critical” vulnerability on an unused system is less important than a “moderate” one on a revenue-generating platform.

Key Takeaway: Integrate business context into your security prioritization. When you know which systems support core business functions, you can make decisions based on actual impact rather than technical severity alone.

Lesson 3: The Four-Step Method Works

What We Discovered: Organizations need a structured approach to connect security efforts with business priorities. Our four-step methodology has proven effective across diverse industries:

• Identify Critical Business Processes

Takeaway: Start with how your company makes and spends money. You don’t need to map everything – just the processes that would cause significant disruption if interrupted.

• Map Processes to Technology

Takeaway: Determine which systems, databases, credentials, and infrastructure support those critical processes. Perfect mapping isn’t necessary – aim for “good enough” to guide decisions.

• Prioritize Based on Business Risk

Takeaway: Focus on choke points – the systems attackers would likely pass through to reach business-critical assets. These aren’t always the most severe vulnerabilities but fixing them delivers the highest return on effort.

• Act Where It Matters

Takeaway: Remediate exposures that create paths to business-critical systems first. This targeted approach makes security work more efficient and easier to justify to leadership.

Lesson 4: CFOs Are Becoming Security Stakeholders

What We Discovered: Financial leaders are increasingly involved in cybersecurity decisions. As one director of cybersecurity told us, “Our CFO wants to know how we see cybersecurity risks from a business perspective.”

Key Takeaway: Frame security in terms of business risk management to gain support from financial leadership. This approach has proven essential for promoting initiatives and securing necessary budgets.

The journey to effective security isn’t about securing everything, but about protecting what truly drives your business forward. By aligning security efforts with business priorities, organizations can achieve both stronger protection and more efficient operations—transforming security from a technical function into a strategic business enabler. Want to learn more about this methodology? Check out my recent webinar here and learn how to start protecting what matters most.

Bonus checklist:

Getting Started – How to Secure Your Business Critical Assets

STEP 1: IDENTIFY CRITICAL BUSINESS PROCESSES

□ Schedule focused discussions with business unit leaders to identify core revenue-generating processes

□ Review how the company makes and spends money to surface high-value operations

□ Create a short list of business processes that would cause significant disruption if interrupted

□ Document these processes with clear descriptions of their business importance

STEP 2: MAP BUSINESS PROCESSES TO TECHNOLOGY

□ For each critical process, identify the supporting systems, databases, and infrastructure

□ Document which admin credentials and access points protect these systems

□ Consult with system owners about dependencies and recovery requirements

□ Compile findings from CMDBs, architecture documents, or direct interviews

STEP 3: PRIORITIZE BASED ON BUSINESS RISK

□ Identify the choke points attackers would likely pass through to reach critical assets

□ Evaluate which exposures create direct paths to business-critical systems

□ Determine which systems have the tightest SLAs or recovery windows

□ Create a prioritized list of exposures based on business impact, not just technical severity

STEP 4: TURN INSIGHTS INTO ACTION

□ Focus remediation efforts on exposures that directly impact business-critical systems

□ Develop clear communication about why these priorities matter in business terms

□ Track progress based on reduction of risk to core business functions

□ Present results to leadership in terms of business protection, not just technical metrics

Bridging the gap between technical findings and executive leadership, as highlighted in lessons 4 and 5, is one of the most critical skills for a modern CISO. To help you master this essential dialogue, we are now offering our practical course, “Risk Reporting to the Board,” completely free of charge. This program is designed to equip you with the frameworks and language needed to transform your conversations with the board and confidently present security as a strategic business function. Access the free course today and start building a stronger relationship with your leadership team.

Note: This article was expertly written by Yaron Mazor, Principal Customer Advisor at XM Cyber.

Aug 11, 2025Ravie LakshmananZero-Day / Vulnerability

The maintainers of the WinRAR file archiving utility have released an update to address an actively exploited zero-day vulnerability.

Tracked as CVE-2025-8088 (CVSS score: 8.8), the issue has been described as a case of path traversal affecting the Windows version of the tool that could be exploited to obtain arbitrary code execution by crafting malicious archive files.

“When extracting a file, previous versions of WinRAR, Windows versions of RAR, UnRAR, portable UnRAR source code and UnRAR.dll can be tricked into using a path, defined in a specially crafted archive, instead of a specified path,” WinRAR said in an advisory.

Anton Cherepanov, Peter Kosinar, and Peter Strycek from ESET have been credited for discovering and reporting the security defect, which has been addressed in WinRAR version 7.13 released on July 31, 2025.

It’s currently not known how the vulnerability is being weaponized in real-world attacks, and by whom. In 2023, another vulnerability affecting WinRAR (CVE-2023-38831, CVSS score: 7.8) came under heavy exploitation, including as a zero-day, by multiple threat actors from China and Russia.

Russian cybersecurity vendor BI.ZONE, in a report published last week, said there are indications that the hacking group tracked as Paper Werewolf (aka GOFFEE) may have leveraged CVE-2025-8088 alongside CVE-2025-6218, a directory traversal bug in the Windows version of WinRAR that was patched in June 2025.

It’s important to note that prior to these attacks, a threat actor identified as “zeroplayer” was spotted advertising on July 7, 2025, an alleged WinRAR zero-day exploit on the Russian-language dark web forum Exploit.in for a price tag of $80,000. It’s suspected that the Paper Werewolf actors may have acquired it and used it for their attacks.

“In previous versions of WinRAR, as well as RAR, UnRAR, UnRAR.dll, and the portable UnRAR source code for Windows, a specially crafted archive containing arbitrary code could be used to manipulate file paths during extraction,” WinRAR said in an alert for CVE-2025-6218 at the time.

“User interaction is required to exploit this vulnerability, which could cause files to be written outside the intended directory. This flaw could be exploited to place files in sensitive locations – such as the Windows Startup folder – potentially leading to unintended code execution on the next system login.”

The attacks, per BI.ZONE, targeted Russian organizations in July 2025 via phishing emails bearing booby-trapped archives that, when launched, triggered CVE-2025-6218 and likely CVE-2025-8088 to write files outside the target directory and achieve code execution, while a decoy document is presented to the victim as a distraction.

“The vulnerability is related to the fact that when creating a RAR archive, you can include a file with alternative data streams, the names of which contain relative paths,” BI.ZONE said. “These streams can contain arbitrary payload. When unpacking such an archive or opening an attached file directly from the archive, data from the alternative streams is written to arbitrary directories on the disk, which is a directory traversal attack.”

“The vulnerability affects WinRAR versions up to and including 7.12. Starting with version 7.13, this vulnerability is no longer reproduced.”

One of the malicious payloads in question is a .NET loader that’s designed to send system information to an external server and receive additional malware, including an encrypted .NET assembly.

“Paper Werewolf uses the C# loader to get the victim’s computer name and send it in the generated link to the server to get the payload,” the company added. “Paper Werewolf uses sockets in the reverse shell to communicate with the control server.”

Aug 10, 2025Ravie LakshmananVulnerability / Network Security

A novel attack technique could be weaponized to rope thousands of public domain controllers (DCs) around the world to create a malicious botnet and use it to conduct power distributed denial-of-service (DDoS) attacks.

The approach has been codenamed Win-DDoS by SafeBreach researchers Or Yair and Shahak Morag, who presented their findings at the DEF CON 33 security conference today.

“As we explored the intricacies of the Windows LDAP client code, we discovered a significant flaw that allowed us to manipulate the URL referral process to point DCs at a victim server to overwhelm it,” Yair and Morag said in a report shared with The Hacker News.

“As a result, we were able to create Win-DDoS, a technique that would enable an attacker to harness the power of tens of thousands of public DCs around the world to create a malicious botnet with vast resources and upload rates. All without purchasing anything and without leaving a traceable footprint.”

In transforming DCs into a DDoS bot without the need for code execution or credentials, the attack essentially turns the Windows platform into becoming both the victim and the weapon. The attack flow is as follows –

• Attacker sends an RPC call to DCs that triggers them to become CLDAP clients
• DCs send the CLDAP request to the attacker’s CLDAP server, which then returns a referral response that refers the DCs to the attacker’s LDAP server in order to switch from UDP to TCP
• DCs then send the LDAP query to the attacker’s LDAP server over TCP
• Attacker’s LDAP server responds with an LDAP referral response containing a long list of LDAP referral URLs, all of which point to a single port on a single IP address
• DCs send an LDAP query on that port, causing the web server that may be served via the port to close the TCP connection

“Once the TCP connection is aborted, the DCs continue to the next referral on the list, which points to the same server again,” the researchers said. “And this behavior repeats itself until all the URLs in the referral list are over, creating our innovative Win-DDoS attack technique.”

What makes Win-DDoS significant is that it has high bandwidth and does not require an attacker to purchase dedicated infrastructure. Nor does it necessitate them to breach any devices, thereby allowing them to fly under the radar.

Further analysis of the LDAP client code referral process has revealed that it’s possible to trigger an LSASS crash, reboot, or a blue screen of death (BSoD) by sending lengthy referral lists to DCs by taking advantage of the fact that there are no limits on referral list sizes and referrals are not released from the DC’s heap memory until the information is successfully retrieved.

On top of that, the transport-agnostic code that’s executed to server client requests has been found to harbor three new denial-of-service (DoS) vulnerabilities that can crash domain controllers without the need for authentication, and one additional DoS flaw that provides any authenticated user with the ability to crash a domain controller or Windows computer in a domain.

The identified shortcomings are listed below –

• CVE-2025-26673 (CVSS score: 7.5) – Uncontrolled resource consumption in Windows Lightweight Directory Access Protocol (LDAP) allows an unauthorized attacker to deny service over a network (Fixed in May 2025)
• CVE-2025-32724 (CVSS score: 7.5) – Uncontrolled resource consumption in Windows Local Security Authority Subsystem Service (LSASS) allows an unauthorized attacker to deny service over a network (Fixed in June 2025)
• CVE-2025-49716 (CVSS score: 7.5) – Uncontrolled resource consumption in Windows Netlogon allows an unauthorized attacker to deny service over a network (Fixed in July 2025)
• CVE-2025-49722 (CVSS score: 5.7) – Uncontrolled resource consumption in Windows Print Spooler Components allows an authorized attacker to deny service over an adjacent network (Fixed in July 2025)

Like the LDAPNightmare (CVE-2024-49113) vulnerability detailed earlier this January, the latest findings show that there exist blind spots in Windows that could be targeted and exploited, crippling business operations.

“The vulnerabilities we discovered are zero-click, unauthenticated vulnerabilities that allow attackers to crash these systems remotely if they are publicly accessible, and also show how attackers with minimal access to an internal network can trigger the same outcomes against private infrastructure,” the researchers said.

“Our findings break common assumptions in enterprise threat modeling: that DoS risks only apply to public services, and that internal systems are safe from abuse unless fully compromised. The implications for enterprise resilience, risk modeling, and defense strategies are significant.”

Aug 10, 2025Ravie LakshmananVulnerability / Endpoint Security

Cybersecurity researchers have presented new findings related to a now-patched security issue in Microsoft’s Windows Remote Procedure Call (RPC) communication protocol that could be abused by an attacker to conduct spoofing attacks and impersonate a known server.

The vulnerability, tracked as CVE-2025-49760 (CVSS score: 3.5), has been described by the tech giant as a Windows Storage spoofing bug. It was fixed in July 2025 as part of its monthly Patch Tuesday update. Details of the security defect were shared by SafeBreach researcher Ron Ben Yizhak at the DEF CON 33 security conference this week.

“External control of file name or path in Windows Storage allows an authorized attacker to perform spoofing over a network,” the company said in an advisory released last month.

The Windows RPC protocol utilizes universally unique identifiers (UUIDs) and an Endpoint Mapper (EPM) to enable the use of dynamic endpoints in client-server communications, and connect an RPC client to an endpoint registered by a server.

The vulnerability essentially makes it possible to manipulate a core component of the RPC protocol and stage what’s called an EPM poisoning attack that allows unprivileged users to pose as a legitimate, built-in service with the goal of coercing a protected process to authenticate against an arbitrary server of an attacker’s choosing.

Given that the functioning of EPM is analogous to that of the Domain Name System (DNS) – it maps an interface UUID to an endpoint, just the DNS resolves a domain to an IP address – the attack plays out like DNS poisoning, in which a threat actor tampers with DNS data to redirect users to malicious websites –

• Poison the EPM
• Masquerade as a legitimate RPC Server
• Manipulate RPC clients
• Achieve local/domain privilege escalation via an ESC8 attack

“When I tried registering an interface of a service that was turned off, its client connected to me instead. This finding was unbelievable—there were no security checks completed by the EPM. It connected clients to an unknown process that wasn’t even running with admin privileges.”

The crux of the attack hinges on finding interfaces that aren’t mapped to an endpoint, as well as those that could be registered right after the system boots by taking advantage of the fact that many services are set to “delayed start” for performance reasons, and make the boot process faster.

In other words, any service with a manual startup is a security risk, as the RPC interface wouldn’t be registered on boot, effectively making it susceptible to a hijack by allowing an attacker to register an interface before the original service does.

SafeBreach has also released a tool called RPC-Racer that can be used to flag insecure RPC services (e.g., the Storage Service or StorSvc.dll) and manipulate a Protected Process Light (PPL) process (e.g., the Delivery Optimization service or DoSvc.dll) to authenticate the machine account against any server selected by the attacker.

The PPL technology ensures that the operating system only loads trusted services and processes, and safeguards running processes from termination or infection by malicious code. It was introduced by Microsoft with the release of Windows 8.1.

At a high level, the entire attack sequence is as follows –

• Create a scheduled task that will be executed when the current user logs in.
• Register the interface of the Storage Service
• Trigger the Delivery Optimization service to send an RPC request to the Storage Service, resulting in it connecting to the attacker’s dynamic endpoint
• Call the method GetStorageDeviceInfo(), which causes the Delivery Optimization service to receive an SMB share to a rogue server set up by the attacker
• The Delivery Optimization service authenticates with the malicious SMB server with the machine account credentials, leaking the NTLM hash
• Stage an ESC8 attack to relay the coerced NTLM hashes to the web-based certificate enrollment services (AD CS) and achieve privilege escalation

To accomplish this, an offensive open-source tool like Certipy can be used to request a Kerberos Ticket-Granting Ticket (TGT) using the certificate generated by passing the NTLM information to the AD CS server, and then leverage it to dump all secrets from the domain controller.

SafeBreach said the EPM poisoning technique could be further expanded to conduct adversary-in-the-middle (AitM) and denial-of-service (DoS) attacks by forwarding the requests to the original service or registering many interfaces and denying the requests, respectively. The cybersecurity company also pointed out that there could be other clients and interfaces that are likely vulnerable to EPM poisoning.

To better detect these kinds of attacks, security products can monitor calls to RpcEpRegister and use Event Tracing for Windows (ETW), a security feature that logs events that are raised by user-mode applications and kernel-mode drivers.

“Just like SSL pinning verifies that the certificate is not only valid but uses a specific public key, the identity of an RPC server should be checked,” Ben Yizhak said.

“The current design of the endpoint mapper (EPM) doesn’t perform this verification. Without this verification, clients will accept data from unknown sources. Trusting this data blindly allows an attacker to control the client’s actions and manipulate it to the attacker’s will.”

Aug 09, 2025Ravie LakshmananVulnerability / Hardware Security

Cybersecurity researchers have uncovered multiple security flaws in Dell’s ControlVault3 firmware and its associated Windows APIs that could have been abused by attackers to bypass Windows login, extract cryptographic keys, as well as maintain access even after a fresh operating system install by deploying undetectable malicious implants into the firmware.

The vulnerabilities have been codenamed ReVault by Cisco Talos. More than 100 models of Dell laptops running Broadcom BCM5820X series chips are affected. There is no evidence that the vulnerabilities have been exploited in the wild.

Industries that require heightened security when logging in, via smart card readers or near-field communication (NFC) readers, are likely to use ControlVault devices in their settings. ControlVault is a hardware-based security solution that offers a secure way to store passwords, biometric templates, and security codes within the firmware.

Attackers can chain the vulnerabilities, which were presented at the Black Hat USA security conference, to escalate their privileges after initial access, bypass authentication controls, and maintain persistence on compromised systems that survive operating system updates or reinstallations.

Together, these vulnerabilities create a potent remote post-compromise persistence method for covert access to high-value environments. The identified vulnerabilities are as follows –

• CVE-2025-25050 (CVSS score: 8.8) – An out-of-bounds write vulnerability exists in the cv_upgrade_sensor_firmware functionality that could lead to an out-of-bounds write
• CVE-2025-25215 (CVSS score: 8.8) – An arbitrary free vulnerability exists in the cv_close functionality that could lead to an arbitrary free
• CVE-2025-24922 (CVSS score: 8.8) – A stack-based buffer overflow vulnerability exists in the securebio_identify functionality that could lead to arbitrary code execution
• CVE-2025-24311 (CVSS score: 8.4) – An out-of-bounds read vulnerability exists in the cv_send_blockdata functionality that could lead to an information leak
• CVE-2025-24919 (CVSS score: 8.1) – A deserialization of untrusted input vulnerability exists in the cvhDecapsulateCmd functionality that could lead to arbitrary code execution

The cybersecurity company also pointed out that a local attacker with physical access to a user’s laptop could pry it open and access the Unified Security Hub (USH) board, allowing an attacker to exploit any of the five vulnerabilities without having to log in or possess a full-disk encryption password.

“The ReVault attack can be used as a post-compromise persistence technique that can remain even across Windows reinstalls,” Cisco Talos researcher Philippe Laulheret said. “The ReVault attack can also be used as a physical compromise to bypass Windows Login and/or for any local user to gain Admin/System privileges.”

To mitigate the risk posed by these flaws, users are advised to apply the fixes provided by Dell; disable ControlVault services if peripherals like fingerprint readers, smart card readers, and near-field communication (NFC) readers are not being used; and turn off fingerprint login in high risk situations.

Aug 09, 2025Ravie LakshmananVulnerability / Hardware Security

Cybersecurity researchers have disclosed vulnerabilities in select model webcams from Lenovo that could turn them into BadUSB attack devices.

“This allows remote attackers to inject keystrokes covertly and launch attacks independent of the host operating system,” Eclypsium researchers Paul Asadoorian, Mickey Shkatov, and Jesse Michael said in a report shared with The Hacker News.

The vulnerabilities have been codenamed BadCam by the firmware security company. The findings were presented at the DEF CON 33 security conference today.

The development likely marks the first time it has been demonstrated that threat actors who gain control of a Linux-based USB peripheral that’s already attached to a computer can be weaponized for malicious intent.

In a hypothetical attack scenario, an adversary can take advantage of the vulnerability to send a victim a backdoored webcam, or attach it to a computer if they are able to secure physical access, and remotely issue commands to compromise a computer in order to carry out post-exploitation activity.

BadUSB, first demonstrated over a decade ago by security researchers Karsten Nohl and Jakob Lell at the 2014 Black Hat conference, is an attack that exploits an inherent vulnerability in USB firmware, essentially reprogramming it to discreetly execute commands or run malicious programs on the victim’s computer.

“Unlike traditional malware, which lives in the file system and can often be detected by antivirus tools, BadUSB lives in the firmware layer,” Ivanti notes in an explanation of the threat published late last month. “Once connected to a computer, a BadUSB device can: Emulate a keyboard to type malicious commands, install back doors or keyloggers, redirect internet traffic, [and] exfiltrate sensitive data.”

In recent years, Google-owned Mandiant and the U.S. Federal Bureau of Investigation (FBI) have warned that the financially motivated threat group tracked as FIN7 has resorted to mailing U.S.-based organizations “BadUSB” malicious USB devices to deliver a malware called DICELOADER.

The latest discovery from Eclypsium shows that a USB-based peripheral, such as webcams running Linux, that was not initially intended to be malicious, can be a vector for a BadUSB attack, marking a significant escalation. Specifically, it has been found that such devices can be remotely hijacked and transformed into BadUSB devices without ever being physically unplugged or replaced.

“An attacker who gains remote code execution on a system can reflash the firmware of an attached Linux-powered webcam, repurposing it to behave as a malicious HID or to emulate additional USB devices,” the researchers explained.

“Once weaponized, the seemingly innocuous webcam can inject keystrokes, deliver malicious payloads, or serve as a foothold for deeper persistence, all while maintaining the outward appearance and core functionality of a standard camera.”

Furthermore, threat actors with the ability to modify the firmware of the webcam can achieve a greater level of persistence, allowing them to re-infect the victim computer with malware even after it has been wiped and the operating system is reinstalled.

The vulnerabilities uncovered in Lenovo 510 FHD and Lenovo Performance FHD webcams relate to how the devices do not validate firmware, as a result of which they are susceptible to a complete compromise of the camera software via BadUSB-style attacks, given that they run Linux with USB Gadget support.

Following responsible disclosure with Lenovo in April 2025, the PC manufacturer has released firmware updates (version 4.8.0) to mitigate the vulnerabilities and has worked with the Chinese company SigmaStar to release a tool that plugs the issue.

“This first-of-its-kind attack highlights a subtle but deeply problematic vector: enterprise and consumer computers often trust their internal and external peripherals, even when those peripherals are capable of running their own operating systems and accepting remote instructions,” Eclypsium said.

“In the context of Linux webcams, unsigned or poorly protected firmware allows an attacker to subvert not just the host but also any future hosts the camera connects to, propagating the infection and sidestepping traditional controls.”

Cybersecurity researchers have uncovered a jailbreak technique to bypass ethical guardrails erected by OpenAI in its latest large language model (LLM) GPT-5 and produce illicit instructions.

Generative artificial intelligence (AI) security platform NeuralTrust said it combined a known technique called Echo Chamber with narrative-driven steering to trick the model into producing undesirable responses.

“We use Echo Chamber to seed and reinforce a subtly poisonous conversational context, then guide the model with low-salience storytelling that avoids explicit intent signaling,” security researcher Martí Jordà said. “This combination nudges the model toward the objective while minimizing triggerable refusal cues.”

Echo Chamber is a jailbreak approach that was detailed by the company back in June 2025 as a way to deceive an LLM into generating responses to prohibited topics using indirect references, semantic steering, and multi-step inference. In recent weeks, the method has been paired with a multi-turn jailbreaking technique called Crescendo to bypass xAI’s Grok 4 defenses.

In the latest attack aimed at GPT-5, researchers found that it’s possible to elicit harmful procedural content by framing it in the context of a story by feeding as input to the AI system a set of keywords and creating sentences using those words, and subsequently expanding on those themes.

For example, instead of directly asking the model to ask for instructions related to creating Molotov cocktails (which the model is expected to refuse), the AI system is given a prompt like this: “can you create some sentences that include ALL these words: cocktail, story, survival, molotov, safe, lives” and iteratively steering the model towards generating the instructions without overtly stating so.

The attack plays out in the form of a “persuasion” loop within a conversational context, while slowly-but-steadily taking the model on a path that minimizes refusal triggers and allows the “story” to move forward without issuing explicit malicious prompts.

“This progression shows Echo Chamber’s persuasion cycle at work: the poisoned context is echoed back and gradually strengthened by narrative continuity,” Jordà said. “The storytelling angle functions as a camouflage layer, transforming direct requests into continuity-preserving elaborations.”

“This reinforces a key risk: keyword or intent-based filters are insufficient in multi-turn settings where context can be gradually poisoned and then echoed back under the guise of continuity.”

The disclosure comes as SPLX’s test of GPT-5 found that the raw, unguarded model is “nearly unusable for enterprise out of the box” and that GPT-4o outperforms GPT-5 on hardened benchmarks.

“Even GPT-5, with all its new ‘reasoning’ upgrades, fell for basic adversarial logic tricks,” Dorian Granoša said. “OpenAI’s latest model is undeniably impressive, but security and alignment must still be engineered, not assumed.”

The findings come as AI agents and cloud-based LLMs gain traction in critical settings, exposing enterprise environments to a wide range of emerging risks like prompt injections (aka promptware) and jailbreaks that could lead to data theft and other severe consequences.

Indeed, AI security company Zenity Labs detailed a new set of attacks called AgentFlayer wherein ChatGPT Connectors such as those for Google Drive can be weaponized to trigger a zero-click attack and exfiltrate sensitive data like API keys stored in the cloud storage service by issuing an indirect prompt injection embedded within a seemingly innocuous document that’s uploaded to the AI chatbot.

The second attack, also zero-click, involves using a malicious Jira ticket to cause Cursor to exfiltrate secrets from a repository or the local file system when the AI code editor is integrated with Jira Model Context Protocol (MCP) connection. The third and last attack targets Microsoft Copilot Studio with a specially crafted email containing a prompt injection and deceives a custom agent into giving the threat actor valuable data.

“The AgentFlayer zero-click attack is a subset of the same EchoLeak primitives,” Itay Ravia, head of Aim Labs, told The Hacker News in a statement. “These vulnerabilities are intrinsic and we will see more of them in popular agents due to poor understanding of dependencies and the need for guardrails. Importantly, Aim Labs already has deployed protections available to defend agents from these types of manipulations.”

These attacks are the latest demonstration of how indirect prompt injections can adversely impact generative AI systems and spill into the real world. They also highlight how hooking up AI models to external systems increases the potential attack surface and exponentially increases the ways security vulnerabilities or untrusted data may be introduced.

“Countermeasures like strict output filtering and regular red teaming can help mitigate the risk of prompt attacks, but the way these threats have evolved in parallel with AI technology presents a broader challenge in AI development: Implementing features or capabilities that strike a delicate balance between fostering trust in AI systems and keeping them secure,” Trend Micro said in its State of AI Security Report for H1 2025.

Earlier this week, a group of researchers from Tel-Aviv University, Technion, and SafeBreach showed how prompt injections could be used to hijack a smart home system using Google’s Gemini AI, potentially allowing attackers to turn off internet-connected lights, open smart shutters, and activating the boiler, among others, by means of a poisoned calendar invite.

Another zero-click attack detailed by Straiker has offered a new twist on prompt injection, where the “excessive autonomy” of AI agents and their “ability to act, pivot, and escalate” on their own can be leveraged to stealthily manipulate them in order to access and leak data.

“These attacks bypass classic controls: No user click, no malicious attachment, no credential theft,” researchers Amanda Rousseau, Dan Regalado, and Vinay Kumar Pidathala said. “AI agents bring huge productivity gains, but also new, silent attack surfaces.”