AI Bait and Stolen Wallets: How Hackers Turn Chatbots and Google Ads Against macOS Users

It starts with a simple Google search: “How do I clear disk space on my Mac?” But for unsuspecting users, that innocent question could be the first step into a sophisticated trap. Today’s hackers are repurposing the very tools that millions trust-AI chatbots like ChatGPT and Grok, and Google’s own search ads-to cloak their malware in plain sight. The prize? Your saved passwords, browser data, and, above all, your cryptocurrency wallets.

Fast Facts

• Attackers embed malicious instructions in shared ChatGPT and Grok conversations, boosting visibility with Google Ads.
• The scam tricks users into running Terminal commands that install the notorious Atomic macOS Stealer (AMOS).
• AMOS operators target over 100 Chrome crypto wallet extensions and pair theft with phishing for brands like Ledger, Trezor, and Exodus.
• Malware often arrives as “signed” and notarized apps, bypassing many of Apple’s built-in security checks.
• Affiliate-style revenue sharing is driving rapid expansion and innovation in macOS cybercrime.

AI, Ads, and the New Face of Social Engineering

The latest AMOS campaign lays bare a chilling trend: cybercriminals are blending the credibility of AI-generated advice with the reach of Google Ads. According to researchers at Malwarebytes and Flare, the infection chain begins when a user’s search for everyday Mac help leads to a seemingly helpful ChatGPT or Grok conversation. But the chat, surfaced via a sponsored ad, contains step-by-step instructions that direct users to copy and paste a Terminal command.

That command, often disguised as a fix, decodes and downloads a malicious script-no traditional app installation required. The twist? The instructions are hosted on legitimate AI domains and pushed to the top of search results, making them nearly indistinguishable from real tech support.

Meanwhile, AMOS isn’t just after passwords. Its operators have their sights locked on cryptocurrency. By targeting Chrome wallet extensions and running phishing campaigns for major crypto brands, they’re siphoning off digital assets at scale. Behind the scenes, a criminal “affiliate” economy flourishes, with stealer operators offering partners a 50/50 split of stolen cryptocurrency and unfettered access to other stolen data logs.

Even Apple’s own security checks are being gamed. Recent malware samples arrived as notarized, signed applications-passing Gatekeeper and lulling users into a false sense of safety. Large decoy files and environmental checks further hinder analysis and detection.

Defensive Red Flags

• Be wary of any web page or chat advising you to run Terminal commands, especially those that download and execute scripts.
• Signed apps requesting unusual permissions or passwords should raise suspicion.
• Monitor for unexpected connections from non-financial apps to blockchain networks.

All signs point to a future where macOS malware campaigns blend into the digital landscape, using the very platforms we trust to erase our natural caution. The era of obvious warning signs is fading; vigilance is now the user’s last line of defense.

Conclusion

As cybercrime syndicates sharpen their tactics, macOS users-and defenders-face an uncomfortable truth: trust is the new attack surface. When malware wears the mask of legitimate AI advice and Apple-approved software, only skepticism and careful scrutiny can keep threats at bay. In the age of AI, every search result could be a wolf in sheep’s clothing.

WIKICROOK

• Infostealer: An infostealer is malware designed to steal sensitive data-like passwords, credit cards, or documents-from infected computers without the user's knowledge.
• Notarization: Notarization is Apple's process of scanning and approving apps for safety, helping prevent malware from running on Macs before distribution.
• Gatekeeper: Gatekeeper is a macOS security feature that checks downloaded apps for authenticity and blocks potentially harmful software from running on your Mac.
• Base64: Base64 encodes binary data into ASCII text, enabling safe transmission of files or images over web applications and email systems.
• Phishing: Phishing is a cybercrime where attackers send fake messages to trick users into revealing sensitive data or clicking malicious links.