🔍 Read the full analysis: One Hacker, Many AI Agents: Rethinking The Cost Of A Break-In on ThorstenMeyerAI.com
Get business pricing on tech for your team
- Business-only prices and quantity discounts
- Tax-exempt purchasing
- Multiple users, one account, clear invoices
TL;DR
Reuters reports that an unidentified Chinese-speaking operator used the open-source Cairn platform to direct AI agents at more than 1,200 websites, with break-ins reported at more than 100 companies. NorForge’s cost estimate is about $5,000 for the campaign, but the estimate is uncertain and Reuters could not establish how much data, if any, was taken.
An unidentified Chinese-speaking operator used the open-source Cairn platform to direct AI agents at online shops and other websites, with Reuters reporting break-ins at more than 100 companies. The activity, which began around September 25, shows how one operator can coordinate automated probing across many targets, though the amount of data taken and the full impact remain unknown.
According to Reuters and NorForge, the agents tried more than 1,200 sites by mid-October, targeting organizations in the United States, Canada, Australia and Japan. Reuters identified affected companies including farm-equipment maker Deere, appliance maker SharkNinja and engine maker Cummins by reviewing activity recorded in the operator’s Cairn panels and contacting companies. NorForge’s report groups the observed activity into clusters but does not name victims.
NorForge researcher Snorre Fagerland found exposed Cairn control panels while searching for internet-connected dashboards. The firm linked more than 20 panels to likely malicious activity. Its analysis says the largest cluster, labeled AIA0001, recorded more than 1,100 attack projects across 14 panels. NorForge measured successful projects as taking just under 23 hours on average, with one running for more than five days.
NorForge estimates the campaign’s AI costs at roughly $5,000 overall, or about $50 per successful intrusion, while warning that the estimate has considerable uncertainty. The figures are estimates, not an independently confirmed accounting of the operator’s expenses. Reuters could not establish how much information, if any, was taken from affected sites.
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
Automation Lowers the Cost of Persistence
The reported campaign matters because it suggests that automated agents can sustain long, repeated reconnaissance with limited direct attention from an operator. The significance is not that AI has made intrusion effortless or guaranteed success: the evidence describes activity across many sites, but does not establish the full damage or the operator’s identity. NorForge’s cost figures are estimates, and should not be treated as a verified price for a typical breach.
For defenders, the reported targeting also puts attention on online shops, support pages and third-party services, not just a company’s main website. A vulnerable vendor-operated site can create security and response work for a larger organization. If agents generate many attempts and findings, security teams may also face more alerts to sort and verify. The source material does not quantify that workload, but the scale of automated activity makes it a practical concern.
The case also raises a question about accountability: Cairn is an open-source, general-purpose agent platform, and its use in this campaign does not by itself show that the software was created for cybercrime. Investigators and companies still need to distinguish the platform’s design from the operator’s actions and establish what happened at each affected site.
Products Worth Considering
Samsung 27-inch Odyssey OLED G6 QD-OLED QHD Gaming Monitor
LG UltraGear 27-Inch QHD OLED Gaming Monitor, 240Hz, 0.03ms, G-SYNC Compatible, FreeSync Premium Pro, HDMI 2.1
Samsung 27-Inch Odyssey QD-OLED G8 4K 240Hz Gaming Monitor
As an affiliate, we earn on qualifying purchases.
How Researchers Traced Cairn Panels
The investigation followed a separate report published in September by Gambit Security. That firm said it recovered a criminal staging server containing three AI harnesses, including Cairn. Harnesses let AI models carry out tasks through software tools; some platforms provide web dashboards to coordinate their activity. NorForge then searched for exposed Cairn panels and found examples that appeared to be used maliciously.
NorForge reported its findings on October 19, 2026, the same day Reuters published its investigation. The firm described agents sharing notes about discoveries and next steps through a common state board, allowing work to continue across multiple agents. Reuters independently reviewed the activity to identify companies and seek their responses. The two reports provide evidence of the campaign, but they do not establish every target, outcome or item of data involved.
The September Gambit report described a separate agent-driven campaign, not the same operation. Gambit said at least 27 companies were compromised and at least 600,000 card records were taken from two of them. Those findings offer a recent comparison for agent-assisted cybercrime, but they should not be attributed to the Cairn campaign.
“a little under 23 hours”
— NorForge report
Victim Impact Is Still Unclear
The reporting does not establish how much data, if any, was taken from the affected websites, or whether attackers used any access for further activity. Deere told Reuters it had no indication that sensitive information was accessed. SharkNinja said a standalone product-support site was affected and that the threat was contained. Cummins said it was investigating a potential cyberattack on a website operated by a third-party vendor.
The operator’s identity is also unknown. The available reporting describes the person as a Chinese-speaking actor; it does not establish a name, location or state affiliation. No arrest or criminal charge is reported in the source material. The number of companies Reuters identified is not a complete, independently verified count of all affected organizations, and the cost estimate comes from NorForge’s analysis.
Company Investigations Continue
Companies named in the reports will need to establish what systems were accessed, whether information was exposed and whether third-party services were involved. Reuters said Cummins was investigating the potential attack on a vendor-operated website; the source material provides no later findings. Deere and SharkNinja have described their responses, but the full technical scope and any lasting effects are not yet public.
Further reporting may clarify which sites were affected and whether the access was used for data theft or other activity. For now, the best-supported account comes from NorForge’s analysis of exposed control panels and Reuters’ review of the operator’s records and company responses. Neither source establishes the campaign’s complete impact or identifies who directed it.
Key Questions
Who was behind the campaign?
The operator has not been identified. Reporting describes the actor as Chinese-speaking, but does not establish a name, location or affiliation.
How many companies were affected?
Reuters reported break-ins at websites belonging to more than 100 companies. The agents reportedly tried more than 1,200 sites, but the full number of affected organizations is not established.
What is Cairn?
Cairn is a free, open-source platform for coordinating AI agents. Its author says it was not designed for hacking attacks; its reported use in this campaign does not establish that the platform itself was built for cybercrime.
Was sensitive data stolen?
That remains unclear. Reuters could not establish how much data, if any, was taken. Deere said it had no indication that sensitive information was accessed; SharkNinja said it contained the threat at a standalone support site.
How much did the campaign cost?
NorForge estimated roughly $5,000 in AI costs for the campaign, or about $50 per successful intrusion. The firm described the estimate as considerably uncertain; it is not a verified accounting of total costs.
Source: ThorstenMeyerAI.com
