Microsoft has identified an active cryptojacking campaign that exploits AI chatbot interactions to steer users toward malicious download sites. The threat actors leverage chatbot responses to bypass traditional security measures and increase exposure to cryptojacking malware.

The campaign represents a shift in social engineering tactics. Rather than relying solely on search engine manipulation, attackers now weaponize AI chatbots to deliver malware recommendations with apparent legitimacy. Users trust chatbot suggestions, making them vulnerable when the bot directs them to compromised domains hosting cryptojacking code.

Cryptojacking malware runs unauthorized cryptocurrency mining operations on victim machines. The attacks consume system resources, degrade performance, and increase electricity costs. For organizations, compromised infrastructure translates to operational inefficiency and potential data center strain.

The technique exploits a fundamental weakness in chatbot design. These systems generate responses based on training data and may inadvertently recommend malicious sites if attackers poison the underlying information sources or manipulate query responses. Users perceive chatbot recommendations as vetted suggestions, lowering their defenses compared to random web search results.

Microsoft Defender Experts observed the campaign through telemetry across multiple endpoints. The threat actors craft queries designed to prompt chatbot recommendations for specific software, utilities, or tools. The bots respond with links to attacker-controlled domains that mimic legitimate software download pages.

Organizations and individuals should verify software sources independently before downloading. Check official vendor websites rather than relying on chatbot recommendations alone. Keep systems patched and deploy endpoint protection that detects cryptojacking signatures. Monitor CPU usage and network traffic for anomalous patterns indicating unauthorized mining.

The emergence of chatbot-based delivery mechanisms demonstrates how threat actors adapt to new technologies. As AI tools become more integrated into user workflows, attackers will continue exploiting trust in these systems. Security awareness training should now include warnings about chatbot recommendations as a potential attack vector