Exposed Dashboards Reveal AI Hacking Toolkit Hit 1,100+ Targets Across 100+ Companies

Security researchers discovered that the control dashboards for an AI-powered hacking toolkit called Cairn had been left exposed on the open internet, revealing the tool had been used in more than 1,100 attempted intrusions against over 100 companies—including John Deere, SharkNinja, and Cummins—with roughly a 10% success rate.

Key points:

• Cairn appears to automate large portions of the hacking process using AI, letting attackers scale up reconnaissance and intrusion attempts well beyond what a small human team could do manually.

• Investigators suspect the tool was built using a Chinese open-source AI model, specifically pointing to DeepSeek V4.1 Flash as a likely underlying component.

• The exposure was accidental—researchers found the dashboards through routine internet scanning, not through a breach of the attackers' own systems.

A roughly 10% success rate across 1,100+ attempts still adds up to approximately 100+ successful intrusions, making this one of the largest documented AI-driven attack campaigns uncovered to date.

The fact that the attackers' own infrastructure was left exposed is a reminder that AI lowers the skill floor for running large-scale attacks, but doesn't necessarily raise the operational security bar—organizations on both sides of this fight are making basic mistakes.

If your company's name wasn't on the disclosed victim list, it's still worth checking whether any of your vendors or supply-chain partners were named, since Cairn's targets spanned industrial and consumer companies with no obvious common thread.

Why It Matters: This exposure confirms AI is now a major force multiplier for attackers—and that even threat actors themselves make operational security mistakes that defenders can exploit.