The achievement highlights how artificial intelligence is changing vulnerability research by reducing the time needed to discover complex software flaws.

How Kimi K3 Found the Vulnerabilities

The AI was given a simple task: search for memory-related security issues such as buffer overflows and use-after-free bugs in the Redis source code.

To complete the task, Kimi K3 automatically:

• Cloned the Redis source code.
• Performed fuzz testing to trigger unexpected behavior.
• Analyzed application crashes.
• Used debugging tools to identify the root cause.
• Generated proof-of-concept (PoC) exploits in a controlled testing environment.

The entire process was completed without manual intervention, showing how AI can automate complex security research.

Redis Versions Affected

The reported vulnerabilities were found in the following Redis versions:

• Redis 6.2.22
• Redis 7.4.9
• Redis 8.6.4
• Redis 8.8.0

Researchers shared the proof-of-concept findings publicly for validation and further analysis.

Key Vulnerabilities Identified

The AI agent reportedly discovered multiple critical issues, including:

• A double-free vulnerability related to Redis stream consumer groups (CVE-2026-25589).
• A heap overflow vulnerability in the RedisBloom module.

These types of memory corruption vulnerabilities can allow attackers to manipulate application memory and may eventually lead to remote code execution if successfully exploited.

Why This Matters

Finding serious vulnerabilities has traditionally required experienced researchers spending days or even weeks reviewing source code and performing extensive testing.

AI-powered tools are now changing that process by:

• Reducing vulnerability discovery time.
• Automating repetitive security testing.
• Identifying bugs across large codebases much faster.
• Helping security teams strengthen software before attacks occur.

Opportunities and Risks

While AI offers significant advantages for defenders, it also introduces new concerns.

Potential benefits include:

• Faster vulnerability detection.
• Improved software security.
• More efficient penetration testing.
• Better support for security researchers.

However, the same technology could also be used by attackers to accelerate exploit discovery, making timely patching and responsible disclosure more important than ever.

The Future of AI in Cybersecurity

The Kimi K3 demonstration shows that AI is becoming an important tool in modern cybersecurity. As large language models continue to improve, they are expected to assist with vulnerability research, code analysis, and automated security testing at a much larger scale.

Organizations should prepare by:

• Applying security updates quickly.
• Continuously monitoring critical systems.
• Using AI-assisted security tools alongside traditional defenses.
• Strengthening vulnerability management programs.

Conclusion

Although these findings are still being reviewed by the security community, they demonstrate how quickly AI-powered security research is evolving. Kimi K3’s ability to discover complex Redis vulnerabilities in minutes signals a major shift in how software security testing may be performed in the future, making AI an increasingly valuable asset for both cybersecurity researchers and defenders.

The post Kimi K3 AI Discovers Redis RCE Flaws appeared first on First Hackers News.

The issue has sparked discussions about how even privacy-focused technologies can be affected by weaknesses outside their core design.

How Hide My Email Protects Users

As part of Apple’s ongoing efforts to enhance security, the Apple Hide My Email feature continues to evolve, providing users with innovative ways to maintain their privacy online.

As part of Apple’s ongoing efforts to enhance security, the Apple Hide My Email feature continues to evolve, providing users with innovative ways to maintain their privacy online.

Hide My Email allows users to create unique email aliases whenever they sign up for websites or online services. Messages sent to these aliases are automatically forwarded to the user’s real inbox, while the actual email address remains hidden.

This approach helps users reduce spam, limit online tracking, and avoid exposing their primary email address if a service experiences a data breach.

What Went Wrong?

Although the feature successfully generated anonymous email aliases, researchers found that the email delivery process itself could unintentionally reveal the original email address.

Instead of breaking the alias directly, the weakness appeared during the way certain emails were processed. Under specific conditions, backend email handling systems could expose information that linked the alias to the user’s real address.

This meant attackers didn’t have to compromise an Apple account—they only needed to exploit the email processing workflow.

Why This Matters

Privacy tools are designed to separate a user’s identity from the services they use.

If someone can trace an anonymous email alias back to its owner, it becomes much easier to build user profiles, launch targeted phishing campaigns, or connect accounts across multiple platforms.

While the vulnerability did not expose passwords or give attackers access to Apple accounts, it weakened an important layer of privacy that many users relied on.

Apple Responds

After the issue was reported through responsible disclosure, Apple investigated the vulnerability and later released a security update to address it.

The fix prevents the conditions that allowed email aliases to be linked with real email addresses, strengthening the privacy protections offered by Hide My Email.

The update highlights Apple’s continued efforts to improve security across its ecosystem as new vulnerabilities are identified.

Can Older Aliases Still Be a Concern?

Although the vulnerability has now been patched, security experts note that information exposed before the update cannot always be erased.

Some email providers and enterprise mail systems store message logs for operational and security purposes. If an email address was previously revealed during processing, that information may still exist in archived records.

Because of this, users who created aliases before the fix may wish to review important accounts that rely on older aliases.

This incident demonstrates that protecting privacy is about more than building secure features.

Modern email systems involve multiple technologies, including spam filtering, message routing, logging services, and third-party infrastructure. Even when the main application is secure, weaknesses elsewhere in the communication process can introduce unexpected privacy risks.

Organizations developing privacy-focused services must evaluate the entire ecosystem rather than only the application itself.

Best Practices for Users

While Apple has resolved the vulnerability, users should continue following good security practices:

• Keep Apple devices updated with the latest security releases.
• Review important online accounts that use older email aliases.
• Enable Multi-Factor Authentication (MFA) wherever available.
• Be cautious of unexpected emails requesting personal information.
• Regularly monitor important accounts for suspicious activity.

Privacy tools like Hide My Email remain valuable for reducing online tracking and limiting data exposure. However, they work best when combined with strong account security and regular software updates to provide multiple layers of protection.

The post Apple Strengthens Hide My Email Privacy with New Security Fix appeared first on First Hackers News.

The vulnerability affects the copy-on-write (CoW) feature of XFS. It is caused by a race condition that occurs when two O_DIRECT write operations access the same reflink-enabled file at nearly the same time.

How the Vulnerability Works

Normally, when a shared file is modified, XFS creates a new private copy of the data before writing changes. However, during this process, the kernel briefly releases an internal lock while waiting for transaction log space.

If another write request arrives during this short window, it can change the file mapping before the first operation finishes. As a result, the first write uses outdated information and writes data directly to the original disk block instead of a new one.

Since O_DIRECT bypasses the page cache, the corrupted data is written straight to disk, allowing attackers to overwrite protected system files.

Why It Is Dangerous

Qualys demonstrated that a normal local user could exploit the flaw on a default RHEL 10.2 system and remove the root account’s password protection within seconds. After the attack, the system granted passwordless root access.

The attack is difficult to detect because:

• No kernel log entries are generated.
• The changes remain after a reboot.
• Even systems protected by SELinux are vulnerable.
• Existing Linux security protections do not block the attack.

Researchers believe the vulnerability has existed since Linux kernel version 4.11, released in 2017, potentially affecting more than 16 million systems worldwide.

Who Is Affected?

A system is vulnerable if it has:

• Linux kernel 4.11 or later without the security patch.
• An XFS filesystem with reflink=1 enabled.
• A directory writable by an unprivileged user.
• A valuable target file such as a SUID binary or protected system file.

Confirmed affected distributions include:

• RHEL 8, 9 and 10
• CentOS Stream 8, 9 and 10
• Oracle Linux 8, 9 and 10
• Rocky Linux 8, 9 and 10
• AlmaLinux 8, 9 and 10
• CloudLinux 8, 9 and 10
• Amazon Linux 2 and Amazon Linux 2023
• Fedora Server 31 and later

Lower-risk distributions

• Debian
• Ubuntu
• SUSE

These distributions are mainly affected only if XFS with reflink support has been manually configured.

Existing Security Features Cannot Stop It

One of the biggest concerns is that common Linux security protections do not prevent this attack. Technologies such as SELinux, KASLR, SMEP, SMAP, kernel lockdown, and container isolation operate at different layers and cannot stop exploitation of this filesystem flaw.

At present, there is no temporary workaround that completely mitigates the vulnerability. Installing the security update is the only effective solution.

AI Helped Discover the Flaw

The vulnerability was identified through a collaboration between Qualys and Anthropic. Researchers used Anthropic’s Claude Mythos Preview AI model to search for race-condition vulnerabilities similar to the well-known Dirty COW bug.

After identifying the issue, Qualys engineers independently verified the findings, created a proof-of-concept, and responsibly disclosed the vulnerability to Linux maintainers.

Part of a Growing Trend

RefluXFS is one of several major Linux privilege escalation vulnerabilities disclosed during 2026. Other recent discoveries include:

• Copy Fail (CVE-2026-31431)
• Dirty Frag (CVE-2026-43284 and CVE-2026-43500)
• DirtyClone (CVE-2026-43503)

These vulnerabilities highlight a growing trend of attackers exploiting flaws that allow protected files or memory to be modified, ultimately leading to privilege escalation.

What Organizations Should Do

Organizations should patch affected systems as soon as possible, especially internet-facing servers, multi-tenant environments, and shared systems.

Security updates are already available for major enterprise Linux distributions, including RHEL, Oracle Linux, AlmaLinux, Rocky Linux, and Fedora. After installing the update, administrators should perform a full system reboot to ensure the patched kernel is running.

Since there is currently no reliable mitigation other than patching, keeping systems updated is the best defense against the RefluXFS vulnerability.

The post Critical RefluXFS Flaw Threatens Linux Systems appeared first on First Hackers News.

Unlike traditional malware campaigns that depend on phishing emails or fake software downloads, this operation targets both developers and AI assistants by poisoning the AI software supply chain. The technique, referred to as AgentBaiting, allows attackers to exploit the trust placed in public AI repositories and documentation, increasing the chances that malicious projects are recommended and installed.

Researchers identified approximately 7,600 malicious GitHub repositories operated by around 6,600 GitHub accounts. Among them, nearly 1,400 repositories were related to AI tools, agents, or automation workflows, while more than 800 repositories specifically impersonated AI Skills or MCP servers.

The campaign spans both personal and enterprise-focused AI tools, including Gmail automation, WhatsApp integrations, AI image generators, resume parsers, Docker connectors, Jenkins plugins, and Databricks integrations. Many of these repositories closely resemble legitimate open-source projects, making them difficult to distinguish from genuine software.

Researchers also observed more than 14 million GitHub Release downloads across approximately 200 campaign repositories. Popular fake projects attracted hundreds of thousands of downloads, while many additional repositories distributed malicious ZIP files directly within the repository itself, making the campaign’s actual reach significantly larger than publicly available download statistics suggest.

How the Attack Works

At the center of the campaign is SmartLoader, a malware loader previously associated with GitHub-based malware distribution. Attackers either clone legitimate repositories or create convincing fake projects that mimic trusted developer tools. These repositories are supported by professional-looking READMEs, fabricated contributor profiles, fake stars, forks, and other GitHub trust signals designed to increase credibility.

The malware installation process is disguised as a normal setup procedure. Downloaded archives typically contain launcher scripts, a LuaJIT-based runtime, and heavily obfuscated Lua payloads that eventually deploy StealC, a well-known information-stealing malware family.

Researchers noted that earlier SmartLoader campaigns had already abused trojanized MCP servers, including cloned Oura Ring integrations, to compromise developer environments. The latest campaign expands this strategy by targeting the rapidly growing AI ecosystem, where Skills, plugins, and MCP servers often receive broad trust without extensive security validation.

One of the most concerning aspects of the campaign is the use of AgentBaiting. Instead of convincing users through traditional phishing techniques, attackers manipulate AI assistants into recommending malicious repositories during legitimate searches.

Testing demonstrated that AI coding assistants and chat-based tools such as Claude Code, Gemini, and ChatGPT can automatically discover attacker-controlled Skills and MCP servers through public registries and GitHub searches. These AI systems may interpret attacker-written documentation as trustworthy installation guidance and present malicious repositories as recommended solutions.

For example, an AI assistant searching for a free cinematic prompt Skill or a free Walmart MCP server could legitimately encounter a malicious repository that has already been indexed by public registries. Since the repository appears authentic and contains convincing documentation, the AI assistant may recommend it without recognizing the underlying threat.

This represents a significant shift in attacker methodology. Rather than directly targeting users, attackers are increasingly influencing the recommendations generated by autonomous AI systems.

The campaign also abuses public AI capability registries that rely heavily on GitHub repositories as trusted sources. Researchers discovered more than 600 malicious listings across platforms including:

• LobeHub
• Glama
• MCP.so
• MCP Market

Many of these registries automatically republish GitHub READMEs and installation instructions, unintentionally amplifying malicious content and increasing its visibility among developers and AI agents.

Impact and Recommended Mitigations

After execution, SmartLoader establishes persistence on the infected system by creating scheduled tasks within the %LOCALAPPDATA% directory. It then retrieves additional encrypted payloads from GitHub and obtains its active command-and-control (C2) server information through values stored inside a Polygon blockchain smart contract, making infrastructure changes more difficult to detect.

The final payload, StealC, is a mature information stealer capable of collecting a wide range of sensitive information, including:

• Browser credentials
• Saved passwords
• Authentication cookies
• Active web sessions
• Email credentials
• Remote access credentials
• Cryptocurrency wallet data
• SSH keys
• Cloud credentials
• Developer secrets
• Host and system information

The campaign highlights a broader security challenge. AI Skills, MCP servers, plugins, and agent workflows are rapidly becoming part of enterprise environments, yet many traditional security solutions—including SIEM platforms, DLP tools, and vulnerability scanners—do not fully inventory or monitor these AI capabilities. As a result, organizations may have limited visibility into what AI assistants connect to or what permissions these integrations receive.

To reduce the risk of AgentBaiting attacks, organizations should treat AI capabilities as software supply chain components rather than simple productivity tools.

Security teams should consider the following best practices:

• Build internal catalogs of approved AI Skills, MCP servers, and plugins.
• Verify publishers instead of relying on GitHub stars, forks, or registry listings.
• Test newly introduced AI capabilities inside isolated environments.
• Reject AI tools distributed as executable files or ZIP archives without transparent source code.
• Monitor AI agent downloads, Git operations, and MCP configuration changes.
• Maintain an inventory of all approved AI repositories, versions, and file hashes.
• Continuously review AI integrations for newly identified malicious repositories.
• Restrict AI tools from accessing sensitive credentials until they have been properly validated.

The FakeGit campaign demonstrates that the AI software supply chain has become an attractive target for attackers. As organizations continue integrating AI assistants into development and business workflows, security teams must extend their software supply chain protections beyond traditional applications to include AI Skills, MCP servers, plugins, and agent-driven automation. Without stronger validation and monitoring, trusted AI ecosystems could become an increasingly effective channel for malware distribution and credential theft.

The post Fake AI Skills Fuel SmartLoader Malware Attack: A New Threat appeared first on First Hackers News.

Many Kusto Query Language (KQL) detection rules search only for connections where RemoteIPType is set to “Public”. However, some internet-bound connections are recorded differently, which means they may not appear in search results or trigger alerts. This creates a blind spot that attackers could exploit to hide malicious outbound communications.

Why Does This Happen?

The DeviceNetworkEvents table is one of the most important telemetry sources in Microsoft Defender XDR. It provides detailed information about endpoint network activity, including:

• Processes initiating the connection
• Command-line arguments
• User accounts
• Remote IP addresses
• Ports
• URLs
• Protocol information

Security analysts frequently use this data to investigate suspicious HTTPS traffic, DNS requests, remote management tools, and potential command-and-control (C2) communications.

The issue occurs because of the RemoteIPType field. Modern Windows systems often use dual-stack networking, allowing applications to communicate using both IPv4 and IPv6.

When an IPv4 address is stored as an IPv4-mapped IPv6 address, Microsoft Defender XDR may classify it as FourToSixMapping instead of Public.

For example:

• IPv4 address: 8.8.8.8
• Recorded value: ::ffff:8.8.8.8
• RemoteIPType: FourToSixMapping

If a detection rule only searches for:

| where RemoteIPType == "Public"

the connection to 8.8.8.8 will not appear in the results because it has been classified as FourToSixMapping instead of Public.

This behavior was discovered during a purple-team exercise after a covert external C2 channel failed to trigger the expected alert. The investigation revealed that the detection rule unintentionally ignored IPv4-mapped IPv6 traffic because it only filtered for Public IP types.

Impact on Detection

This classification behavior can create a significant detection gap, particularly in organizations that use dual-stack networking.

Detection rules designed to monitor:

• Command-and-control (C2) traffic
• Suspicious outbound HTTPS connections
• Unusual DNS activity
• Remote access tools
• Proxy-bypass techniques

may fail to detect malicious activity if they rely only on RemoteIPType == “Public”.

Another important consideration is the ipv4_is_private() KQL function. Some analysts use this function to identify private IP addresses, but it does not always work correctly with IPv4-mapped IPv6 values.

For example, when the function receives an address such as:

::ffff:8.8.8.8

it may return null instead of true or false.

Since null is neither true nor false in KQL filtering logic, mapped IPv4 addresses can be silently excluded from search results if the IP address is not normalized first.

To avoid this issue, Microsoft recommends including both Public and FourToSixMapping values in detection queries and normalizing mapped IPv4 addresses before applying additional filters.

A recommended query is shown below:

| where RemoteIPType in ("Public", "FourToSixMapping")
| extend RemoteIP = iff(
    RemoteIPType == "FourToSixMapping",
    replace_string(RemoteIP, "::ffff:", ""),
    RemoteIP
)

After normalizing the IP address, analysts can consistently filter private RFC 1918 addresses, loopback addresses, and other non-internet destinations without accidentally excluding legitimate outbound traffic.

Security teams should review existing Microsoft Defender XDR and Microsoft Sentinel detection rules that rely on RemoteIPType == “Public”. Rules designed to detect C2 activity, suspicious outbound HTTPS traffic, unusual DNS requests, remote management tools, and proxy-bypass techniques should be updated to account for both Public and FourToSixMapping values.

By making these changes, organizations can improve detection coverage, reduce false negatives, and ensure that outbound communications are monitored more accurately across both IPv4 and IPv6 environments.

The post Microsoft Defender XDR C2 Detection Gap appeared first on First Hackers News.

The change applies to all Copilot users, including both free and paid subscribers. Since Microsoft does not provide an option to export or download these podcasts, all existing podcast content and shared links will be permanently removed after the feature is discontinued.

Microsoft Ends the Copilot Podcasts Feature

Copilot Podcasts allowed users to turn prompts and documents into AI-generated podcast-style audio. However, the podcasts were stored only inside the Copilot app and could not be downloaded or moved to another platform.

Microsoft confirmed that once the feature is retired, previously created podcasts and any shared links pointing to them will stop working. The company thanked users for their feedback but did not explain why the feature is being removed.

This retirement only affects the consumer Copilot app and does not impact Microsoft 365 subscriptions or other Copilot features included with eligible plans.

What Users and Organizations Should Do

Users and organizations that have shared Copilot podcast links in documents, training materials, websites, or internal communications should review them before August 18, 2026. After the retirement date, those links will no longer be accessible.

Although the original AI-generated audio cannot be exported, users may be able to recreate important podcasts if they still have the original prompts, documents, notes, or transcripts used to generate them.

This announcement also highlights an important lesson for organizations using AI services. Business-critical AI-generated content should not rely solely on cloud-hosted features that do not support data export. Keeping local copies or storing important AI-generated content in approved repositories can help avoid data loss if a service or feature is removed in the future.

Microsoft confirmed that this change only affects the Podcasts feature. Other Copilot capabilities and Microsoft 365 subscription benefits will continue to be available as usual. Users who need additional information can contact Microsoft Support through the official support channels.

The post Microsoft Ends Copilot Podcasts and Access appeared first on First Hackers News.

The vulnerabilities, identified as CVE-2026-15409 and CVE-2026-15410, were observed being exploited before public disclosure. Because of the active attacks, both flaws have been added to CISA’s Known Exploited Vulnerabilities (KEV) catalog, emphasizing the need for immediate action.

Exploitation Chain

The first vulnerability allows an unauthenticated attacker to reach internal services that are normally protected from external access. This provides an entry point to the appliance and enables attackers to execute commands with limited privileges.

The second vulnerability can then be used to escalate those privileges to root by abusing the appliance’s update and rollback process. Once exploited, attackers can run malicious scripts with full administrative permissions, giving them complete control over the system. Researchers also noted that the appliance may reboot after exploitation, which can make incident investigation more difficult.

Affected products include SonicWall SMA 1000 Series devices running vulnerable releases in the 12.4.3 and 12.5.0 software branches. SonicWall has confirmed that the SMA 100 Series and SSL VPN functionality on SonicWall firewalls are not affected.

Security Recommendations

Organizations using affected SonicWall SMA 1000 appliances should install the latest security hotfixes released by SonicWall as soon as possible, as no alternative mitigation is currently available.

In addition to patching, security teams should inspect appliances for signs of compromise, re-image affected systems if necessary, rotate administrator and user credentials, reset multi-factor authentication tokens following a confirmed breach, and continue monitoring VPN infrastructure for suspicious activity.

Applying the latest updates and reviewing exposed appliances promptly will help reduce the risk of compromise from these actively exploited vulnerabilities.

The post Attackers Target SonicWall SMA1000 Zero-Days appeared first on First Hackers News.

The vulnerability is linked to insecure cloud permissions and a built-in command execution feature found in certain Shark vacuum firmware. At the time of disclosure, no official patch had been released.

How the Vulnerability Works

Researchers found that the issue affects internet-connected Shark robot vacuums that communicate through AWS IoT services. By exploiting weak cloud permissions, an attacker who gains access to one vulnerable device could potentially interact with other Shark vacuums connected to the same cloud environment.

The most serious concern is a command execution feature that allows specially crafted messages to run commands on the device. During testing, researchers successfully demonstrated that credentials taken from one Shark vacuum could be used to execute commands on another compatible device.

Some Shark models also include cameras and store household maps for navigation. Researchers warned that a successful attack could allow unauthorized access to:

• Live camera feeds on supported models
• Home mapping data
• Stored Wi-Fi credentials
• Remote control of the vacuum
• Device information and telemetry

The research also identified a large number of active Shark devices communicating with the affected cloud infrastructure, indicating that the potential impact could be significant.

Recommended Security Measures

Until an official fix is released, Shark robot vacuum owners should take steps to reduce their exposure.

Recommended actions include:

• Keep the vacuum firmware updated with the latest available version.
• Monitor SharkNinja for future security updates and advisories.
• Place smart home and IoT devices on a separate Wi-Fi network whenever possible.
• Regularly review your home network for unknown or suspicious devices.
• Remove unused smart devices from your network.
• Change your Wi-Fi password if you suspect unauthorized access.

This vulnerability highlights the growing importance of securing Internet of Things (IoT) devices. As more connected devices become part of everyday life, regular updates, network segmentation, and strong security practices remain essential for protecting home networks and personal data.

The post Critical Flaw Found in Shark Robot Vacuums appeared first on First Hackers News.

The malicious packages were distributed as .NET command-line tools, allowing users to install them using the standard dotnet tool install command. They primarily targeted players of popular games such as Albion Online, GTA5RP, GrandRP, Majestic RP, Lineage 2, Russian Fishing 4, and Throne and Liberty.

After discovering the campaign, researchers reported the packages to NuGet’s security team for removal.

How the Malware Operates

Once installed, the malicious package downloads a second-stage payload called pepesoft.exe from attacker-controlled infrastructure hosted on GitHub Releases and Hugging Face.

The malware includes several techniques to avoid detection and maintain its operation. It uses Google DNS-over-HTTPS to resolve download servers, making it more difficult for traditional DNS-based security controls to block its activity. In many cases, it also attempts to synchronize the system clock before downloading additional components.

After installation, Pepesoft contacts remote servers to retrieve configuration updates and collect information from the infected device. The malware generates a unique hardware identifier and gathers system information such as usernames, computer names, hardware details, network information, public IP addresses, and approximate location.

Some variants also include Telegram bot functionality, allowing attackers to remotely capture screenshots of the victim’s desktop. These screenshots could expose browser sessions, password managers, cryptocurrency wallets, authentication prompts, private conversations, or other sensitive information displayed on the screen.

Security Recommendations

Developers and organizations should review their development environments for suspicious NuGet packages and investigate any unexpected execution of pepesoft.exe.

Recommended security measures include:

• Review installed NuGet packages and remove any unknown or suspicious tools.
• Verify the authenticity of open-source packages before installation.
• Monitor systems for unexpected execution of pepesoft.exe.
• Watch for unusual outbound connections to GitHub, Hugging Face, and other unknown infrastructure.
• Monitor for unexpected DNS-over-HTTPS traffic from development systems.
• Keep endpoint security solutions updated and regularly scan developer workstations.
• Educate developers about software supply chain attacks and package verification.

This campaign highlights the growing risk of software supply chain attacks, where attackers disguise malicious packages as legitimate developer tools. Carefully verifying third-party packages before installation remains one of the most effective ways to reduce this risk.

The post NuGet Packages Deliver Pepesoft Malware appeared first on First Hackers News.

Although the vulnerability cannot be exploited remotely, it poses a significant risk because attackers often use privilege escalation flaws after initially compromising a system.

How the Vulnerability Works

WinFsp is an open-source framework that enables Windows applications to create and use custom file systems. It is commonly integrated into software that provides virtual drives, cloud storage access, and other file-system virtualization features.

According to security researchers, the flaw is caused by a synchronization issue within the WinFsp driver. By exploiting this weakness, an attacker can trigger memory corruption inside the Windows kernel. If the attack is successful, the attacker can elevate their privileges from a standard user account to the highly privileged SYSTEM account.

With SYSTEM privileges, an attacker may be able to install malicious software, disable security tools, create new administrator accounts, modify protected files, or maintain long-term access to the compromised device.

Protecting Windows Systems

The issue has been fixed in the latest WinFsp release, and organizations should ensure all affected systems are updated without delay.

Security teams should also:

• Upgrade to the latest WinFsp version.
• Check whether third-party applications include vulnerable WinFsp components.
• Limit local administrator privileges wherever possible.
• Monitor for unusual driver, service, or privilege escalation activity.
• Investigate unexpected processes interacting with WinFsp.
• Keep Windows and endpoint security solutions fully updated.

Privilege escalation vulnerabilities like this are commonly used as part of multi-stage attacks. While they require an attacker to already have some level of access, they can significantly increase the impact of a compromise by allowing attackers to bypass security controls and gain complete control of a Windows system.

The post WinFsp Security Flaw Enables Privilege Escalation appeared first on First Hackers News.