AIThis post was created with the assistance of artificial intelligence (AI).

🔍 Read the full analysis: How AI Agents Are Changing The Price Of A Website Break-In on ThorstenMeyerAI.com

Before you orderOffer from Amazon

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
Start your free Prime trial Free trial for eligible customers · Cancel anytime
As an affiliate, we earn on qualifying purchases.

TL;DR

A Chinese-speaking operator used Cairn, a free, open-source platform for coordinating AI agents, to probe online shops beginning Sept. 25. Reuters, citing the operator’s activity records, reported break-ins at websites of more than 100 companies; NorForge estimated about $5,000 in AI costs for the campaign, but said the estimate carries considerable uncertainty.

A Chinese-speaking operator used the open-source Cairn platform to direct AI agents at online shops from Sept. 25, with the agents probing more than 1,200 websites and breaking into sites belonging to over 100 companies, according to records reviewed by Reuters and security firm NorForge. The campaign matters because it shows how one operator can automate lengthy reconnaissance across many targets, although the estimated cost per successful intrusion is uncertain and the extent of any data theft is not known.

NorForge researcher Snorre Fagerland found exposed Cairn control panels—web dashboards used to coordinate agents—and monitored their activity. NorForge said it identified more than 20 panels it linked to likely malicious use. Reuters reported that the largest activity cluster, which NorForge labeled AIA0001, recorded more than 1,100 attack projects across 14 panels. The operators’ notes were written in Chinese, but the identity of the person or people behind the campaign has not been established.

NorForge measured successful AIA0001 projects as taking a little under 23 hours from start to goal on average, with one project running for more than five days. It estimated the campaign’s AI costs at roughly $5,000 overall, or about $50 per successful hack, while warning that the estimate had “considerable” uncertainty. Those figures are estimates, not a verified accounting of the operator’s spending or total costs.

Reuters identified companies by reviewing the attacker’s Cairn activity and contacting them. Deere told Reuters it took “necessary remediation actions” and had “no indication that sensitive information was accessed.” SharkNinja said a standalone product-support website was affected and the threat was contained. Cummins said it was investigating “a potential cyber-attack” on a website run by a third-party vendor. Reuters could not establish how much data, if any, was taken. Two smaller businesses acknowledged compromises, according to the source material.

At a glance
reportWhen: Campaign began Sept. 25, 2026; NorForge…
The developmentA security researcher observed AI agents coordinated through Cairn probing more than 1,200 websites, in a campaign Reuters says breached sites belonging to over 100 companies.
Thorsten Meyer AI
Cluster AIA0001 · Sep–Oct 2026

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.

<100Chinese characters of instructions from the operator (Reuters)
1,200+sites the agents had tried by mid-October (Reuters)
100+companies whose websites were broken into, per Reuters' count from the attacker's records
~23haverage from start to goal for successful projects (NorForge)

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

  1. Shop websites are ranked by traffic and checked automatically in a browser
  2. Each target becomes a Cairn project: a starting point and a goal
  3. Agents post facts and next steps on a shared board and build on each other
  4. 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 suggests AI agents can take on repeated, time-consuming steps of website probing while an operator sets goals and monitors results. NorForge’s account describes agents sharing findings and proposed next steps through Cairn, allowing activity to continue across many targets without a person manually directing every action. That does not mean the software independently chose victims or carried out every part of an intrusion; the operator supplied instructions and the platform coordinated the work.

The estimated $50 per successful hack illustrates a possible low marginal cost, but the figure is not a confirmed market price for a break-in. NorForge itself described the estimate as highly uncertain. For companies, the reported targets also draw attention to support websites and third-party vendor systems, which may be less closely monitored than core services but can still provide a route into business infrastructure.

Automated activity can also increase the burden on defenders: many attempts may generate alerts that require investigation, while successful access may be sought for later use or resale. NorForge interpreted the campaign’s goal as gaining persistent access to e-commerce and payment infrastructure for possible later monetization. That is the firm’s assessment of intent, not confirmation that every compromised site was used that way.

Products Worth Considering

Amazon Kindle Paperwhite Signature Edition (12th Generation, 2024), 32 GB, Metallic Black

Amazon Kindle Paperwhite Signature Edition (12th Generation, 2024), 32 GB, Metallic Black

LG 34GX90SA-W 34-inch Ultragear WQHD OLED Curved Gaming Monitor

LG 34GX90SA-W 34-inch Ultragear WQHD OLED Curved Gaming Monitor

As an affiliate, we earn on qualifying purchases.

Exposed Cairn Panels Revealed Activity

The investigation developed after Gambit Security reported a separate agent-driven campaign in September. Its report described at least 27 companies compromised and at least 600,000 card records taken from two of them. That campaign was distinct from the activity NorForge tracked; the figures should not be combined or treated as proof that the later targets suffered the same data theft.

NorForge searched for internet-visible Cairn dashboards, some of which lacked password protection. The firm said it found more than 20 panels associated with likely malicious activity and grouped the observed behavior into three clusters. Its report, published Oct. 19, 2026, named no victims. Reuters published its investigation the same day and said it identified company targets by independently reviewing Cairn activity and seeking comment.

Cairn is a general-purpose, open-source agent-coordination platform, not a hacking tool by design. Its author, who uses the name le1xia0, said on X that Cairn “was not designed for hacking attacks” and that harmful use was not the author’s intention, Reuters reported. The platform’s availability does not by itself establish who operated the observed panels or how they were configured.

““Most need authorization to access. Most.””

— NorForge report

Data Theft and Operator Identity Unclear

The operator’s identity remains unknown. The evidence described in the reports points to a Chinese-speaking actor, based on the language used in activity records; it does not establish the person’s nationality or location. The available reporting also does not establish an arrest, charge, or public identification of a suspect.

It is not clear how much information, if any, was taken from the affected websites. Deere said it had no indication that sensitive information was accessed, while SharkNinja said the threat was contained. Cummins described its investigation as ongoing. Reuters could not determine the amount of data involved across the reported incidents, and the available material does not establish that every listed company experienced the same kind of access or impact.

The campaign’s overall cost is also unresolved. NorForge’s approximate $5,000 estimate depends on assumptions about AI usage and costs; the source material does not provide an independently verified bill. Nor does the reporting establish whether access was later used, sold, or exploited beyond the activity observed in Cairn.

Company Investigations and Panel Security

The immediate next steps are for affected companies and vendors to complete their investigations, determine whether information was accessed, and disclose any confirmed impact. Cummins said it was investigating a potential attack; the reports provide no final findings or timetable. Companies should also review exposed management dashboards and the security of standalone support sites and third-party services, while avoiding assumptions that an exposed panel alone proves a successful breach.

Further reporting may clarify which sites were accessed, whether any information was taken, and what happened to the access afterward. NorForge’s monitoring of Cairn panels gave researchers an unusually direct view of the campaign’s recorded activity, but it cannot answer every question about systems beyond those records. For now, the central finding is the scale of automated probing reported by Reuters and NorForge, alongside significant gaps about the consequences for individual companies.

Key Questions

What is Cairn?

Cairn is an open-source platform for coordinating AI agents through shared findings and proposed next steps. Its author said it was not designed for hacking, according to Reuters.

How many websites were targeted?

Reuters reported that the agents tried more than 1,200 sites by mid-October. NorForge counted more than 1,100 attack projects across 14 panels in the largest activity cluster; those are separate measures of the observed activity.

How much did the campaign cost?

NorForge estimated about $5,000 in AI costs for the campaign, or roughly $50 per successful hack. The firm said uncertainty around the estimate was considerable, and it is not a verified total cost.

Was customer data stolen?

That has not been established for the campaign described here. Reuters could not determine how much data, if any, was taken. Deere reported no indication of sensitive information access, while the other company investigations had different reported statuses.

Has the operator been identified?

No. The reporting describes a Chinese-speaking operator, but does not identify a person, establish nationality, or report an arrest or charge.

Source: ThorstenMeyerAI.com

You May Also Like

GitHub’s AI Agent Surge Moves The Pull Request Bottleneck To Review

GitHub says agents started 110 million pull requests in September. Its new tools focus on review, approval and the costs of checking AI-written code.

Enhancing AI Performance Through Fine-Tuning Multi-Vector Sentence Embeddings

Sentence Transformers v6.0 introduces MultiVectorEncoder for domain-specific retrieval, showing promising results in medical search, pending independent validation.

15 Best Smart Plugs With Voice Control To Automate Your Home In 2026

A 2026 smart plug roundup compares voice assistant support, Matter, outlet design, energy monitoring and local control.

Resident Evil 4 (GameCube) – Complete Byte-identical Decompilation To C/C++

A complete decompilation of Resident Evil 4 for GameCube has been achieved, resulting in byte-identical C/C++ source code, marking a significant milestone in game reverse engineering.