🔍 Read the full analysis: A Hacker And A Swarm Of AI Agents: Exploring The $50 Break-In on ThorstenMeyerAI.com
Get the latest gadgets delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
TL;DR
A Chinese-speaking operator used Cairn, a free, open-source AI-agent platform, to probe online shops beginning Sept. 25, according to NorForge and Reuters. Reuters counted break-ins at websites of more than 100 companies from the operator’s records, but the amount of data accessed or taken has not been established.
A Chinese-speaking operator used the open-source platform Cairn to coordinate AI agents probing online shops from Sept. 25 through mid-October, according to security firm NorForge and Reuters. Reuters counted break-ins at websites of more than 100 companies from the attacker’s own records, including Deere, SharkNinja and Cummins; how much information, if any, was taken remains unknown.
NorForge researcher Snorre Fagerland found exposed Cairn control panels while searching the internet for the platform’s dashboards. The panels allowed him to observe activity associated with what NorForge assessed as likely malicious use. The firm identified more than 20 panels it linked to that activity, and Reuters reported that new panels continued to appear.
NorForge grouped the activity into three clusters. Its largest, labeled AIA0001, used Chinese-language task instructions. NorForge counted more than 1,100 attack projects across 14 panels in that cluster, aimed mainly at the United States, Japan and Australia. Reuters separately reviewed activity in the attacker’s Cairn records and identified company websites, then contacted the organizations.
NorForge measured successful AIA0001 projects as taking a little under 23 hours from start to goal. It estimated the campaign’s AI costs at about $5,000, or roughly $50 per successful intrusion, while warning that the estimate carries considerable uncertainty. These figures describe the researcher’s estimate, not a verified accounting of the operator’s costs.
One operator, a swarm of agents
The campaign in numbers
What the researcher and Reuters report about the busiest group NorForge found. Counts come from the attackers’ own exposed panels; cost and success figures are estimates.
The price tag estimates
NorForge / Fagerland, “uncertainty considerable”
- ~$5 per task on average, including tasks that fail early
- $100+ for long projects with more than 1,000 intents and facts
- ~$5,000 in AI costs for the campaign, about $50 per successful hack
- ~10% success rate, NorForge’s “reasonable assumption”
How the swarm runs
Simplified, no attack details
- Shop websites are ranked by traffic and checked automatically in a browser
- Each target becomes a Cairn project: a starting point and a goal
- Agents post facts and next steps on a shared board and build on each other
- Goal: a hidden way back in to shop and payment systems, for later
- 22 SepGambit reportGambitCairn found among three AI harnesses on a criminal’s staging server (separate campaign)
- 25 SepCampaign startsReutersA Chinese-speaking operator turns agents on shops in the US, Canada, Australia and Japan
- ≈ 2 OctOpen panels foundReutersFagerland starts finding Cairn panels in malicious use, about two dozen
- mid-Oct1,200+ sites triedReutersBreaks into websites of 100+ companies, incl. Deere, SharkNinja, Cummins
- 19 OctReport and investigationNorForge · ReutersNorForge publishes “Inside Cairn”; Reuters publishes its investigation
Sources: NorForge, Inside Cairn (19 Oct 2026); Reuters (19 Oct 2026); Gambit Security (22 Sep 2026). Numbers as published, same rounding as the article.
Powered by Thorsten Meyer AI
Low-Cost Automation Raises Defense Pressures
The case shows how an operator can use agents to sustain repeated, automated probing across many targets without manually directing every step. The reported roughly $50 estimate per successful intrusion suggests that the cost of persistent attempts may be low, though NorForge says the calculation is uncertain and it does not establish the campaign’s overall profitability.
For companies, the reported targets include online shops and related web infrastructure, not only prominent corporate systems. Reuters’ reporting and company statements point to the need to examine standalone support sites and third-party-operated websites as well as core services. The case also presents a monitoring challenge: automated activity can generate many findings for defenders to review, while attribution to a human operator remains limited.
The report does not show that AI agents independently chose the campaign’s goals or that they can reliably break into systems without human setup. It documents the use of agent software under an operator’s direction. The distinction matters: the technology may expand the scale of activity, but responsibility and intent cannot be assigned to the software itself on the evidence described.
Products Worth Considering
Apple iPad Air (2025) 13‑inch Wi‑Fi 256 GB – Space Gray (M3 chip)
Bose SoundLink Flex (2nd Gen) Portable Bluetooth Speaker
EcoFlow DELTA 3 Classic Portable Power Station, 1024Wh, 1800W
As an affiliate, we earn on qualifying purchases.
How Cairn Panels Exposed the Campaign
The investigation followed a September report from Gambit Security, which described a separate agent-driven campaign. Gambit said it found three AI “harnesses,” including Cairn, on a criminal staging server. It reported at least 27 companies compromised and at least 600,000 card records taken from two of them. That was a separate operation; the figures should not be attributed to the campaign NorForge and Reuters examined.
NorForge then searched for publicly reachable Cairn dashboards. Its report says most panels require authorization, but some were accessible without a password. From around Oct. 2, Fagerland identified about two dozen panels associated with malicious activity, according to Reuters. NorForge published its analysis on Oct. 19, the same day Reuters published its investigation.
Cairn is a general-purpose, open-source system for coordinating AI agents, rather than a tool NorForge describes as purpose-built for hacking. Agents can share a record of findings and intended next steps, allowing work to continue across multiple software processes. That coordination model helps explain how the campaign could run for hours or days, but it does not by itself establish what data was reached or extracted.
““Most need authorization to access. Most.””
— NorForge report
Data Access and Operator Identity
The operator’s identity is not known; the reporting describes the actor as Chinese-speaking. The available account does not establish the person’s location, identity or whether others were involved. It also reports no arrest or criminal charge connected to this campaign.
Reuters could not establish how much data, if any, was taken. Deere said it had no indication that sensitive information was accessed. SharkNinja told Reuters that a standalone product-support site was affected and that the threat had been contained. Cummins said it was investigating a potential cyberattack on a website operated by a third-party vendor. Those statements differ in scope and do not settle what information may have been exposed across all reported victims.
The $5,000 campaign-cost figure and approximate $50-per-intrusion figure are NorForge estimates, not audited expenditures. The reported number of break-ins is based on attacker records reviewed by Reuters; NorForge’s own published report did not name the victims. The reporting does not provide a complete, independently verified accounting of every affected organization or the consequences for each site.
Company Reviews and Further Verification
Companies named in the reporting may continue investigating affected websites, including services run by third-party vendors, and determining whether any data was accessed. Cummins said it was investigating; Deere and SharkNinja described remediation or containment to Reuters. The source material does not give a timetable for final findings or announce further public disclosures.
NorForge’s exposed-panel monitoring supplied a view into the campaign, while Reuters independently reviewed the attacker’s activity records and sought responses from companies. Further confirmation would depend on company investigations, additional technical evidence and any law-enforcement findings. No arrest, charge or public identification of the operator is reported in the material available as of Oct. 30.
Key Questions
What happened in the reported campaign?
A Chinese-speaking operator used Cairn to coordinate AI agents probing online shops beginning Sept. 25. Reuters counted break-ins at websites of more than 100 companies from the operator’s records.
Which companies were named?
Reuters identified Deere, SharkNinja and Cummins among the companies whose websites were involved. Their statements described remediation, containment or an ongoing investigation, but did not establish the same impact for each organization.
Was customer or company data stolen?
That has not been established. Reuters could not determine how much data, if any, was taken. Deere said it had no indication sensitive information was accessed; Cummins said it was investigating a potential attack on a third-party vendor’s website.
Was Cairn built as a hacking tool?
NorForge describes Cairn as a general-purpose, open-source agent platform. Its author said it was not designed for hacking attacks and that harmful use was not their intention. The platform’s general design does not change the reported use in this campaign.
How reliable is the $50-per-intrusion estimate?
It is a NorForge estimate, based on its calculation of about $5,000 in AI costs for the campaign and the reported successful intrusions. NorForge called the uncertainty considerable, so the figure is not a verified cost or a measure of damages.
Source: ThorstenMeyerAI.com
