ClickFix Social Engineering Surges as Attackers Expand to AI Lures and macOS

Security researchers report that ClickFix has become a major initial-access technique, relying on social engineering rather than software exploits to trick users into pasting attacker-provided commands into Windows Run or macOS Terminal. ESET said detections rose 108% from H2 2025 to H1 2026, while RH-ISAC described a broader 517% increase from late 2024 into the first half of 2025 and noted that Microsoft saw the method in 47% of initial-access cases handled by Defender Experts. Attackers have expanded ClickFix beyond Windows to macOS, delivered it through compromised WordPress sites, and improved lures with fake BSOD screens, frozen document viewers, and service-specific error messages.

The technique is now used by both cybercriminals and state-backed groups, with RH-ISAC linking adoption to Sandworm, APT28, MuddyWater, and Kimsuky and citing Sandworm activity against organizations in Ukraine, including at least one compromise involving FreakyPoll malware. Researchers also observed ClickFix delivering malware such as ACR Stealer, OkoBot, TELEPUZ, ClickLock Stealer, and DriveSurge, while newer variants including CrashFix, FileFix, PromptFix, and ConsentFix show continued evolution. ESET additionally identified an "AI-fix" variant that abuses legitimate domains tied to Anthropic Artifact, OpenAI Canvas, and Microsoft Copilot Pages to display fake troubleshooting content for nonexistent AI problems, underscoring how attackers are blending trusted platforms, AI branding, and increasingly industrialized delivery infrastructure.