DeepLoad logs keystrokes, buries details behind reams of AI-generated code, and re-infect hosts days after being blocked, according to ReliaQuest.

The massive amount of junk code that hides the malware's logic from security scans was almost certainly generated by AI, researchers say.

Researchers at ReliaQuest warn of persistent malware campaign targeting enterprise credentials

The massive amount of junk code that hides the malware's logic from security scans was almost certainly generated by AI, researchers say.

A newly discovered malware named DeepLoad is targeting enterprise environments, turning a single user action into persistent, credential-stealing access that survives reboots and outlasts standard cleanup efforts. What sets this campaign apart is how every stage of the attack was deliberately built to defeat the security controls that most organizations already depend on. DeepLoad arrives through ClickFix, where attackers display a fake browser error page and instruct employees to paste a PowerShell command into their Windows Run dialog to “fix” it. That one command creates a scheduled task that re-executes the loader on every reboot and uses mshta.exe, a legitimate Windows utility, to fetch an obfuscated payload from attacker-controlled infrastructure. The staging domains were already serving malicious content within 22 minutes of going live, giving response teams very little time to act. ReliaQuest analysts and researchers identified this campaign while investigating active enterprise compromises. Their findings showed the full attack chain was built to outpace manual response from the very start. Credential theft begins before the main chain finishes, and the malware spread to USB drives within ten minutes of infection, making the first host unlikely to be the only impacted system. The immediate business risk is real. DeepLoad drops a credential stealer called filemanager.exe — named to blend into any process list — that runs on its own command-and-control channel and steals data even if the primary loader is blocked. A malicious browser extension captures passwords and session tokens as users type them, persisting across sessions until removed. The malware also wrote over 40 disguised installer files to connected USB drives, including fake shortcuts for Chrome, Firefox, and AnyDesk, each ready to trigger a full infection on any machine they touch. Standard cleanup alone is not enough. A hidden WMI event subscription planted during the initial compromise sits outside standard remediation workflows, leaving the host ready to reinfect itself with no user action required. In one confirmed case, that subscription fired three days after the host appeared clean and silently dropped filemanager.exe back into the user’s Downloads folder. AI-Powered Evasion and Process Injection DeepLoad avoids detection at every layer, making it hard to catch with traditional security tools. Its PowerShell loader is padded with thousands of meaningless variable assignments that make the script appear busy without performing any real work. The actual logic — a short XOR decryption routine — sits at the bottom and decrypts shellcode in memory, so no decoded payload touches disk. ReliaQuest researchers assessed with high confidence that AI generated this obfuscation layer, meaning new variants can be rebuilt and redeployed quickly before defenders have time to adjust detection coverage. Once running, the loader uses PowerShell’s Add-Type feature to compile a fresh C# injector on the fly, producing a randomly named DLL that signature-based tools cannot match. The malware then selects a trusted Windows process to inject into — on investigated hosts, it chose LockAppHost.exe, the Windows lock screen process. Since the LockAppHost.exe does not typically initiate outbound connections, most security tools are not configured to monitor it. Through asynchronous procedure call (APC) injection, the loader places shellcode into that process’s memory and triggers execution on resume, leaving no decoded payload on disk. Security teams should enable PowerShell Script Block Logging, since it captures decoded runtime commands and cuts through obfuscation. All WMI event subscriptions on affected hosts must be explicitly audited and cleared before any machine returns to production, as a surviving subscription can re-execute the attack days after cleanup. Every credential reachable from a confirmed infected host — saved passwords, session tokens, and active accounts — must be rotated immediately. All USB drives connected to affected endpoints should be audited before reuse. Browser extensions outside approved IT deployment paths must be removed from affected systems. Endpoint monitoring should shift from file-based scanning to behavioral, runtime detection using EDR telemetry and memory scanning. Follow us on  Google News ,  LinkedIn , and  X  to Get More Instant Updates ,  Set CSN as a Preferred Source in  Google . The post New DeepLoad Malware Uses ClickFix and AI-Generated Evasion to Breach Enterprise Networks appeared first on Cyber Security News .

Researchers at ReliaQuest identified a credential-stealing malware campaign dubbed "DeepLoad" that leverages AI to obfuscate its operations at every stage of the attack lifecycle. The malware logs keystrokes, conceals its code behind layers of AI-generated obfuscation, and can reinfect compromised hosts days after initial detection and blocking. No specific dates, victim organizations, or technical indicators (e.g., CVE IDs) were disclosed in the findings. The campaign demonstrates advanced evasion techniques, including dynamic code generation to bypass static analysis tools. The primary impact involves persistent credential theft and potential lateral movement within infected networks.

DeepLoad logs keystrokes, buries details behind reams of AI-generated code, and re-infect hosts days after being blocked, according to ReliaQuest.

A new campaign has leveraged the ClickFix social engineering tactic as a way to distribute a previously undocumented malware loader referred to as DeepLoad. "It likely uses AI-assisted obfuscation and process injection to evade static scanning, while credential theft starts immediately and captures passwords and sessions even if the primary loader is blocked," ReliaQuest researchers Thassanai

DeepLoad logs keystrokes, buries details behind reams of AI-generated code, and re-infect hosts days after being blocked, according to ReliaQuest.

Introduction: The Silent Evolution of Cyber Threats Cybersecurity threats are no longer just about brute force or simple phishing tricks. A new generation of malware is emerging, one that blends automation, intelligence, and deception into a nearly invisible weapon. The recently discovered “DeepLoad” malware represents a chilling shift in how attacks are designed and executed. By leveraging what researchers believe to be AI-generated code, this malware doesn’t just infiltrate systems, it adapts, hides, and persists in ways that challenge even advanced security defenses.

Pesquisadores da agência de cibersegurança ReliaQuest descobriram como um novo malware consegue roubar credenciais das vítimas. Chamado de DeepLoad, esse agente malicioso consegue infectar a máquina das vítimas por meio da técnica de invasão do ClickFix e já preocupa os especialistas. Esse malware foi inicialmente descoberto em fevereiro, mas só agora os pesquisadores começaram a entender seu modo de operação. Uma vez dentro dos dispositivos, a aplicação consegue apagar seus rastros em plataformas como o PowerShell e usa funções diretas do Windows para não ser detectado.

The DeepLoad malware leverages the ClickFix delivery method and possibly AI-generated evasion to bypass defenses, escalating enterprise cybersecurity risks.

DeepLoad malware uses ClickFix and WMI persistence techniques to steal browser credentials, enabling attackers to maintain access and exfiltrate sensitive data.

DeepLoad logs keystrokes, buries details behind reams of AI-generated code, and re-infect hosts days after being blocked, according to ReliaQuest. 

Windows 11’s security situation continues to deteriorate. Users and businesses, especially the latter, are facing two risky developments: first, a new AI-powered malware threat, and second, additional emergency updates from Microsoft fixing critical security vulnerabilities. This combination highlights just how rapidly the threat landscape in the Windows environment is changing these days. ## AI malware uses new tricks At the center is a new piece of malware called “DeepLoad,” which differs from conventional malware: instead of delivering suspicious files to the computer, it relies on a so-called “fileless” attack method. Specifically, users are tricked into entering seemingly harmless commands into Command Prompt or PowerShell. It’s through this action that the infection is actually triggered—and traditional antivirus scanners, which primarily react to known files, often fail to detect it. Once the system has been compromised, malware can establish a permanent foothold and communicate with the attackers’ servers via legitimate Windows tools. The primary goal is to steal login credentials, particularly in corporate environments. ## AI is changing the threat landscape For you as a user, this means one thing above all: traditional protection mechanisms are increasingly reaching their limits. AI-powered malware can dynamically adapt its code, making it significantly harder to detect. At the same time, the time between a security vulnerability being discovered and the first attacks is shrinking. For home users, the risk is currently lower than in corporate environments. Nevertheless, even everyday attacks are increasingly relying on deception rather than technology. ## What you should do now Even though many current attacks specifically target businesses, there are some basic measures you can start taking right now: * Install Windows updates as soon as possible. * Never execute any commands in PowerShell or Command Prompt if you don’t know _exactly_ what they’re doing 100%. * Be suspicious of instructions found online or in emails—especially in unsolicited emails, which is a red flag of a phishing scam. * Use up-to-date security software as an additional layer of protection. If you don’t have security software, start with PCWorld’s picks for the best antivirus apps on Windows. ## Emergency Windows 11 updates In parallel, Microsoft already released emergency security updates in mid-March. These primarily affect Windows 11 Enterprise versions such as 24H2 and 25H2, as well as the LTSC variant. Several critical vulnerabilities in Routing and Remote Access Service (RRAS) have been patched. Attackers could exploit these flaws to execute malicious code remotely and take complete control of a system. In some scenarios, simply connecting to a compromised server was enough for an attack to succeed. ## Microsoft Office also affected The current issues are part of a wider trend. On Patch Tuesday in March, Microsoft patched more than 80 vulnerabilities, including critical flaws in Excel and other Office applications. In some cases, simply opening the preview pane in Outlook was enough to execute malicious code. Furthermore, initial examples show that AI features such as Copilot can also create new attack vectors—for instance, when data is inadvertently passed on via automated processes. _By the way: If you’re using Windows 11 Home, you’re missing out on the many benefits of Windows 11 Pro. To learn more, see our comparison of Windows 11 Home and Pro. If you want to upgrade, snag it for cheap in the PCWorld Software Store: now just $59 instead of $99._

 Pressure on Windows 11 security grows - driven by emerging AI-powered malware alongside unpatched flaws threatening companies and everyday users alike. The pace of change in digital threats becomes clearer through recent incidents, especially within large organizational networks. DeepLoad sits at the heart of recent cybersecurity worries. This particular threat skips typical download tactics altogether.  Instead of dropping files, it operates without any - earning its "fileless" label. Users themselves become part of the breach process. By following deceptive prompts, they run benign-looking instructions in system utilities such as Command Prompt. Once executed, those inputs quietly trigger malicious activity behind the scenes. Since nothing gets written to disk, standard virus scanners often miss what's happening.  Detection becomes difficult when there’s no file footprint to flag. After running, the malware stays active by embedding itself into system processes while reaching out to remote servers through standard Windows tools. Because it targets confidential information like passwords, its presence poses serious risks inside business environments. What makes it harder to detect is how it blends malicious activity with normal operating routines. Security teams may overlook it during routine checks due to this camouflage technique.  Artificial intelligence makes existing threats more dangerous. Because AI-driven malware adjusts on the fly, it slips past standard detection systems. As a result, security tools struggle to keep up. With each change the malware makes, response times shrink. The gap between finding a flaw and facing an attack grows narrower by the hour. Meanwhile, security patches have been rolled out by Microsoft to fix numerous high-risk weaknesses.  Affected are various business-focused builds of Windows 11 - both recent iterations and extended support variants. One major concern involves defects within the Routing and Remote Access Service (RRAS), where exploitation might let threat actors run harmful software from a distance. Full administrative access to compromised machines becomes possible through these gaps. Not just isolated systems feel the impact.  That last Patch Tuesday, Microsoft fixed over eighty security gaps in its programs - problems hiding even inside tools such as Excel and Outlook. Opening an attachment wasn’t needed; sometimes, just looking at it could activate harmful code, showing how dangerous these weaknesses really are. Experts warn that even emerging AI tools, such as Microsoft Copilot, could introduce new risks if not properly secured, particularly when sensitive data is handled automatically.  Though companies face the most attacks, regular individuals can still be affected. When new patches arrive, it helps to apply them without delay - timing often matters more than assumed. Opening unknown scripts carries risk; many breaches begin there. Unexpected requests, especially those demanding immediate steps, deserve extra skepticism.  Change is shaping a new kind of digital danger - cleverer, slyer, built to exploit how people act just as much as system flaws. One moment it mimics trust; the next, it slips through unnoticed.