🔍 Read the full analysis: How To Think About AI Agents And The Cost Of Website Attacks on ThorstenMeyerAI.com
Get the little things that make your day 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
Reuters reports that an unidentified Chinese-speaking operator used Cairn, a free, open-source AI-agent platform, to target more than 1,200 websites from late September to mid-October. Reuters counted website break-ins at more than 100 companies using the operator’s records; the data taken, if any, remains unclear. NorForge estimates the campaign’s AI costs at about $5,000, but says that figure carries considerable uncertainty.
An unidentified Chinese-speaking operator used the open-source Cairn platform to coordinate AI agents against online shops beginning Sept. 25, according to a NorForge report and a Reuters investigation. Reuters counted break-ins at websites of more than 100 companies among more than 1,200 targeted sites, though it remains unclear whether attackers accessed or took sensitive data.
NorForge researcher Snorre Fagerland found exposed Cairn control panels while scanning internet-indexed services. NorForge linked more than 20 panels to likely malicious activity. From around Oct. 2, Fagerland tracked roughly two dozen panels being used in attacks, with new ones appearing. NorForge published its report on Oct. 19, the same day Reuters published its investigation.
The largest activity cluster, named AIA0001 by NorForge, used Chinese-language instructions. NorForge counted more than 1,100 attack projects across 14 panels, aimed largely at the United States, Japan and Australia. The broader campaign described in the source material also targeted sites in Canada. Reuters independently reviewed the attacker’s Cairn activity to identify affected companies; NorForge’s report did not name victims.
NorForge says successful AIA0001 projects took a little under 23 hours on average to reach their goal. It estimates about $5,000 in AI costs for the campaign, or roughly $50 per successful intrusion, while stressing that the estimate has “considerable” uncertainty. Those figures describe NorForge’s estimate, not an independently verified accounting of the operator’s expenses.
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, Wider Exposure
The reported campaign matters because it suggests an operator can delegate repeated website probing to software agents rather than guide every step manually. NorForge’s observations point to long-running, coordinated activity: agents shared findings and next steps through Cairn, while the operator could provide directions without constantly supervising each task. The evidence does not establish that agents can compromise any site, or that this method is cheaper than every alternative.
For companies, the reported targets include not only prominent manufacturers but also the web services around their businesses. Reuters said Deere, SharkNinja and Cummins were among the companies whose websites were identified in the investigation. A support site or a site run by a vendor may be less visible than a company’s main storefront, but it can still expose customers or business systems if compromised. Reuters could not establish how much information, if any, was taken in this campaign.
The operational burden may also extend beyond the initial intrusion. If automated systems generate many simultaneous reports or probing attempts, security teams may have to sort legitimate activity from malicious traffic while checking overlooked sites and third-party services. NorForge’s account raises questions about how organizations monitor these surfaces, but the available reporting does not quantify resulting costs or demonstrate a particular defensive failure across the named companies.
Products Worth Considering
Withings BPM Connect Smart Blood Pressure Monitor
Withings BPM Vision Blood Pressure Monitor with Wi-Fi and Bluetooth
Garmin Venu 3 Fitness Smartwatch, 45 mm, Black
As an affiliate, we earn on qualifying purchases.
How Cairn Panels Exposed Activity
The reporting began with security research into exposed Cairn dashboards. Fagerland searched for internet-visible control panels and found some without password protection. Cairn is an open-source platform for coordinating agents; it was not presented in the source material as a hacking product. NorForge describes it as a general system that works through a series of steps toward a specified goal.
In the campaign NorForge tracked, a software feed reportedly ranked online shops by traffic and checked them in a browser before passing targets to agents. The agents recorded findings as “facts” and proposed further actions as “intents,” allowing later agents to build on earlier work. NorForge’s interpretation is that the activity sought persistent access to e-commerce or payment infrastructure for possible later monetization. That is the researchers’ assessment of the apparent goal, not proof that access was later sold or exploited.
The author of Cairn, who uses the name le1xia0, told Reuters the platform was not designed for hacking and that harmful use was not the author’s intention. The source material also describes a separate September report by Gambit Security: it documented an agent-driven campaign involving at least 27 companies and at least 600,000 card records taken from two of them. That separate incident provides context, but it is not evidence that the same operator or platform was responsible for both campaigns.
““a little under 23 hours””
— NorForge report
Data Access and Operator Identity
The operator’s identity is unknown. The reporting describes a Chinese-speaking actor, but does not establish who was behind the activity, where the person was located, or whether anyone has been arrested or charged. The source material reports no arrest or criminal charge.
It is also unclear how much data, if any, attackers accessed or removed. Deere told Reuters it had no indication that sensitive information was accessed; SharkNinja said a standalone product-support site was hit and the threat was contained; and Cummins said it was investigating a potential attack on a vendor-operated website. Reuters could not establish the amount of data taken across the campaign. NorForge’s cost figure is an estimate, and the basis for a definitive total remains uncertain.
Company Reviews and Security Follow-Up
The companies’ response and investigations are the immediate next milestones. Cummins said it was investigating the potential vendor-site attack. Deere reported remediation actions, while SharkNinja said it had contained the threat to a standalone product-support site. The available reporting does not give a timetable for further findings or say whether additional affected organizations will publicly confirm incidents.
NorForge’s findings also leave a practical follow-up for security teams: review exposed control panels, online shop infrastructure, support sites and vendor-operated web properties. That is a defensive implication of the reported attack pattern, not a claim that every organization with those services was targeted. Further reporting may clarify the extent of access, whether any data was taken, and how the campaign’s cost estimate compares with the operator’s actual spending.
Key Questions
What happened in the reported campaign?
An unidentified Chinese-speaking operator used Cairn to coordinate AI agents targeting online shops from late September to mid-October 2026. Reuters counted website break-ins at more than 100 companies among more than 1,200 targeted sites, based on records from the attacker’s Cairn activity.
Who was behind the attacks?
The operator has not been identified. The reporting characterizes the actor as Chinese-speaking; it does not establish the person’s identity or location. The source material reports no arrest or charge.
Did the attackers take customer or company data?
That remains unclear across the campaign. Reuters could not establish how much data, if any, was taken. Deere said it had no indication that sensitive information was accessed; the companies’ statements do not resolve what may have happened at every affected site.
How much did the AI-agent campaign cost?
NorForge estimates roughly $5,000 in AI costs for the campaign, or about $50 per successful intrusion. The firm described the estimate as carrying “considerable” uncertainty; it is not a confirmed accounting from the operator.
Was Cairn built as a hacking tool?
No such purpose is established in the reporting. NorForge describes Cairn as a general-purpose, open-source agent-coordination platform. Its author, le1xia0, told Reuters the software was not designed for hacking attacks and that harmful use was not the author’s intention.
Source: ThorstenMeyerAI.com
