Skip to main content

News · Ransomware

How to Read a Ransomware Leak-Site Claim: Maersk, WhiteArmor and the $5 Million Demand

On 11 November 2026 Maersk confirmed unauthorized access and data extraction from a legacy system inherited through an acquisition. WhiteArmor, a group that appeared in October, says it took about 100 GB and wants $5 million. Here is what each side has actually said, what a leak-site claim can and cannot prove, and what a customer or supplier of a named company should do while the facts settle.

isMalicious TeamisMalicious Team
Published 13 min read
On this page11

A ransomware leak-site claim proves one thing: a group wants a company's name next to its own. It does not prove how much data was taken, what that data is, when it was taken, or even that it belongs to the company named. The Maersk case this month shows the difference clearly. Maersk has confirmed unauthorized access and data extraction from a legacy system inherited through an acquisition. WhiteArmor, the group that listed Maersk on its leak site, claims about 100 GB of logistics data and a $5 million ransom. The first is a fact; the second, so far, is a claim. This article separates the two, then turns the case into a method you can reuse the next time a partner of yours appears on a leak site.

What Maersk confirmed, in its own words

On 11 November 2026, A.P. Moller - Maersk published an advisory on its newsroom. The core of it:

A.P. Moller – Maersk has identified an IT security incident involving unauthorised access and data extraction from one of our local legacy systems from a previous acquisition.

Our findings so far strongly suggest no sensitive data has been disclosed. The incident is isolated to a local system, and we have found no indication that there has been any impact on other Maersk systems or customer and partner systems that connect to Maersk through integrations such as APIs and EDI.

The advisory adds that Maersk does not anticipate disruption to operations for customers, that the investigation is ongoing, that it will inform impacted parties if it identifies a need to, and that customers should "treat any unexpected communications claiming to relate to this matter with caution and verify them through your usual Maersk contact."

What the advisory does not say matters as much. It does not name the system, the acquisition, the type of data extracted, the date of the intrusion or the group responsible. It does not mention WhiteArmor at all.

What WhiteArmor claims

WhiteArmor posted Maersk on its leak site over the weekend of 7 and 8 November, according to BleepingComputer and BankInfoSecurity. As those outlets report the post, the group claims:

  • about 100 GB taken from the "OCF/LF Logistics platform";
  • nearly 30,000 driver records, plaintext passwords, 33 SUPER_ADMIN accounts, signed delivery receipts, purchase orders and advance shipping notices, and 27,842 mobile device logs with phone numbers, email addresses and IMEI numbers;
  • a demand of $5 million in bitcoin within 30 days, sized against Maersk's earnings with the line "Five million dollars is roughly four hours of that EBITDA."

WhiteArmor told BankInfoSecurity that Maersk had not contacted it. Neither outlet says it verified the data, and Maersk did not answer BleepingComputer's follow-up questions before publication.

The LF Logistics connection is plausible on its face: Maersk completed its acquisition of LF Logistics, a Hong Kong-based contract logistics company, on 31 August 2022. But "a legacy system from a previous acquisition" is Maersk's phrase, and "LF Logistics" is the group's. Maersk has not connected the two.

Point Maersk, 11 November WhiteArmor's leak site, as reported
Unauthorized access Confirmed Claimed
Data taken "Data extraction" confirmed, type not given About 100 GB, detailed list of record types
Affected system A local legacy system from an acquisition The OCF/LF Logistics platform
Sensitive data "Strongly suggest" none disclosed Plaintext passwords and admin accounts
Other systems and integrations No indication of impact Not addressed
Operations No disruption anticipated Not addressed
Who did it Not named WhiteArmor
Date of intrusion Not given Not given
Ransom Not mentioned $5 million in bitcoin, 30 days

The two accounts overlap on one point: data left a Maersk system. They diverge on whether any of it was sensitive. "Disclosed" in Maersk's advisory is not defined; it can mean published, or exposed to the intruder. A claim of plaintext passwords would, if true, be sensitive under any definition. That gap will close only when Maersk updates its statement or someone independent examines a sample. None of the reports we read, up to 15 November, describes either.

Ransomware claim: what to verify, what to do Decision graphic in three stages. A leak-site claim states a company name, a volume, record types and a ransom with a deadline: the attacker’s assertion. Four questions to verify: has the named company confirmed an incident; has its statement or an independent analysis of a sample established which data was taken; did the affected system hold your data or your staff’s credentials; were indicators of compromise published. Matching actions: treat the claim as unverified and read the company’s exact words; assume your data may be included until the scope is known; rotate credentials, check password reuse and ask the company in writing; hunt the indicators in your logs, then block them. In every case, verify payment and pickup changes through a known contact and check sender domains. Ransomware claims · A method Claim → what to verify → what to do 01 Claim 02 What to verify 03 What to do Leak-site post Company name Volume: “100 GB” Record types Ransom and deadline The attacker’sassertion, chosenfor pressure Has the companyconfirmed an incident? No: unverified claim.Yes: read its exact wordsand note the date. Is the scope established,by its statement or anindependent sample analysis? Not yet: assume your datamay be included. Did the system hold yourdata or your staff’scredentials? Rotate credentials, checkreuse, ask the companyin writing. Were indicators ofcompromise published? Hunt them in your logs,then block them. Whatever the answers: verify payment and pickup changes through a known contact,and check the sender domain of every message that cites the incident.

What a leak-site claim proves, and what it does not

A listing is a press release written by the attacker. It reliably tells you that the group has decided to put pressure on a named company in public. Often, though not always, it also means the group had some access to something connected to that company. Beyond that, every detail is the attacker's assertion:

  • Volume and contents are stated by the party with the most interest in inflating them.
  • Attribution to the company can be wrong. On 23 June 2024 LockBit claimed it had breached the US Federal Reserve and held 33 TB of "Americans' banking secrets". When the files appeared, threat-monitoring firm HackManac found they appeared to come from Evolve Bank & Trust, and Evolve confirmed it was investigating an incident involving "a known cybercriminal organization", as BleepingComputer reported.
  • Freshness is not guaranteed. Check Point Research found that many datasets published by the FunkSec group in late 2024 were recycled from earlier hacktivist leaks, which raised doubts about the authenticity of its disclosures.
  • The intrusion date is almost never on the post.

Our own WhiteArmor page puts it in one line under its victim count: "A claim is not proof of a compromise."

Four dates, not one

A ransomware incident has at least four dates, and they are routinely confused:

  1. Intrusion: when the attacker got in and took data. Neither Maersk nor WhiteArmor has given it.
  2. Discovery: when the victim noticed. Maersk says its Cyber Security team launched an investigation once it discovered the issue, without giving a date.
  3. Claim: when the group posts. For Maersk, the weekend of 7 and 8 November.
  4. Disclosure: when the victim says something in public. For Maersk, 11 November.

The claim is the only date the attacker controls, and it is chosen for pressure, not accuracy. Legal clocks start elsewhere. Under the GDPR, a controller notifies its supervisory authority within 72 hours of becoming aware of a personal data breach. Under NIS2, an essential or important entity sends an early warning within 24 hours of becoming aware of a significant incident, a notification within 72 hours and a final report within a month. A company registered with the US Securities and Exchange Commission files a Form 8-K under Item 1.05 within four business days of determining that an incident is material. Most of these notifications go to authorities, not to the public, so a quiet victim is not necessarily a victim that has not reported. Which of these regimes applies to Maersk's incident is for Maersk to say; it has not.

Why groups exaggerate

Inflation is not a side effect of extortion; it is part of the method.

  • Pressure. A longer list of record types and a precise count ("27,842 mobile device logs") read as proof even when no sample backs them. The "four hours of EBITDA" line frames the ransom as cheap.
  • Visibility. A new group needs a reputation to attract affiliates and make its next demand credible. WhiteArmor is new: the claims our page draws from ransomware.live start on 23 October 2026, and by 15 November it counted 22.
  • Consequences that do not wait for proof. WhiteArmor listed J.B. Hunt on 23 October with a claim of 140,838 employee records. Former employees filed suits seeking class action status within days, citing the group's figures, and one of them states those figures are unverified, according to the Arkansas Democrat-Gazette. J.B. Hunt said it was investigating "a cybersecurity incident involving unauthorized access" limited to "an isolated portion" of its systems.
  • Deletion promises. A deadline implies that payment ends the matter. It does not. In 2020 incident response firm Coveware documented groups that re-extorted victims who had paid, published data that had been paid for, and showed fake files as proof of deletion, and concluded: "The data will not be credibly deleted."

How to verify a claim

The order below runs from the most to the least authoritative.

  1. The named company's own statements, and their updates. Read the exact words: "access", "extraction", "disclosed" and "no indication" each mean something different. Statements change as investigations progress, so note the date you read.
  2. Formal filings and notifications where they are public: an 8-K Item 1.05 for a US-listed company, notification letters sent to affected people, a regulator's published breach list.
  3. Independent analysis of a published sample. Researchers check whether internal identifiers, timestamps and structure match the claimed source, and whether people in the data recognize their records. Troy Hunt describes the method he uses for Have I Been Pwned, and how it exposed fabricated breaches, in his verification write-up. Leave this to researchers and counsel: downloading leaked personal data yourself creates legal exposure and proves little.
  4. Trackers, read for what they count. Leak-site trackers record claims, and they count them differently: one claim per victim and day on our page, other units elsewhere. A tracker tells you that a claim exists and when, not that it is true.
  5. Your own telemetry, if you are connected to the named company: unusual traffic on the integration, failed logins on shared accounts, phishing that cites the incident.

If you work with a company that has been named

Whether the claim turns out to be accurate or not, a public listing hands fraudsters a ready-made pretext to write to the named company's customers and suppliers, which is why Maersk's own advisory warns about unexpected messages. Logistics adds a specific risk: the FBI's April 2026 alert on cyber-enabled cargo theft describes criminals impersonating brokers and carriers by email to redirect loads.

  • Incident-themed messages are suspect by default. "Update on the security incident", "new portal after the breach", "re-verify your account": verify through your usual contact, never through a link or number in the message.
  • Payment and shipment changes need a second channel. The FBI's wording is direct: "Independently verify shipment requests and pickups using secondary methods prior to releasing any loads," and "familiar names or email addresses alone do not confirm authenticity." The same rule applies to a supplier that suddenly asks for new bank details.
  • Credentials on the named system get rotated. If your staff had accounts on a platform the claim names, change those passwords now and look for the same passwords elsewhere. A claim of plaintext passwords, true or not, is exactly what credential stuffing feeds on.
  • Sender domains get checked. The FBI alert lists the usual lookalike patterns: added punctuation, a different top-level domain, an added word, a misspelling. Run unfamiliar sender domains through a domain reputation check and check their domain age: a domain registered the week of the incident deserves the question.
  • Ask in writing what the system held about you. A supplier with a data processing agreement owes you an answer about your data, and the question gets a more precise reply than a general request for news.
  • Indicators get hunted when they exist. Maersk's advisory publishes none, and none of the reports cited here includes any. If an advisory or the company later publishes indicators, search your logs for them and block them. For logistics companies that run their own web portals, Ben Wilkens of the National Motor Freight Traffic Association told FreightWaves that WhiteArmor relies on "very targeted attacks of vulnerable webservers", and that the first step is to ensure EDR is "deployed, active, and monitored on internet-facing web servers", then to investigate unexpected processes launched by web services, suspicious command execution, webshell creation and unusual outbound connections.

For a supplier incident that does turn out to reach your environment, our third-party risk guide covers scoping and containment, and our piece on extortion that adds physical threats covers the response side when your own organization is the one named.

What the isMalicious WhiteArmor page shows

The WhiteArmor group page, read on 15 November 2026, shows:

  • Known victims: 22, defined as distinct claims on its leak site, one per victim and day, as ransomware.live publishes them;
  • first claim 23 October 2026, latest claim 8 November 2026, with 22 claims in the last 30 days against none in the 30 days before;
  • targeted sectors: technology 6, transportation and logistics 5, other 4, retail and consumer 3, manufacturing 2, government 1, healthcare 1;
  • targeted countries: the United States 6, one each for Argentina, China, Colombia, Germany, Denmark, the United Kingdom and the Philippines, and 9 not specified;
  • the Tor hidden service associated with the group, with a warning to open it only from an isolated environment (we do not reproduce it here);
  • a lookup to check whether your domain appears on the group's victim list, and an alert, with an account, when one of your domains is named on a ransomware leak site.

The page counts claims. It does not confirm them, and it does not tell you what was taken. Its use is the one this article argues for: knowing early that a partner has been named, so the verification and the vigilance above start before the invoice-fraud emails arrive. The ransomware tracker lists the other groups on the same model.

What to do now

  1. Record what Maersk has confirmed and what WhiteArmor claims as two separate columns in your notes, with the date you read each, and update them as Maersk publishes more.
  2. If you are a Maersk customer or supplier, brief finance and logistics staff today: incident-themed messages and payment or pickup changes are verified through a known contact.
  3. If your staff held accounts on the platform the claim names, the former LF Logistics systems, rotate those credentials and check for password reuse, without waiting for Maersk to confirm which system was affected.
  4. Check the sender domains of any message that cites the incident, and block lookalikes.
  5. Ask your Maersk contact in writing whether the affected system held your data.
  6. For every future claim against a partner, run the same four questions: confirmed by the company, scope established, your data concerned, indicators published.

Sources

Frequently asked questions

What did Maersk confirm about the WhiteArmor ransomware claim?

In an advisory dated 11 November 2026, A.P. Moller - Maersk said it had identified unauthorized access and data extraction from one of its local legacy systems from a previous acquisition. It said its findings so far strongly suggest no sensitive data has been disclosed, that the incident is isolated to that system, and that it does not anticipate disruption to customer operations. The statement does not name WhiteArmor, the system, the data involved or the date of the intrusion.

What does WhiteArmor claim to have stolen from Maersk?

According to BleepingComputer and BankInfoSecurity, which reported the leak-site post, WhiteArmor claims about 100 GB from the OCF/LF Logistics platform, including driver records, plaintext passwords, administrator accounts, signed delivery receipts and mobile device logs, and demands $5 million in bitcoin within 30 days. None of these details had been independently verified in the reports published up to 15 November 2026.

Does a ransomware leak-site listing prove a company was breached?

No. A listing proves that a group wants the name attached to its brand. It does not establish how much data was taken, what it contains, when it was taken or whether it belongs to the named company. In June 2024 LockBit claimed the US Federal Reserve, and the data it posted turned out to come from Evolve Bank & Trust. Statements by the victim and independent analysis of published samples are what settle the scope.

What should a supplier or customer of a company named on a leak site do?

Treat any message that refers to the incident as suspect, and confirm payment-detail changes and shipment or pickup requests through a contact you already know, never through details given in the message. Rotate passwords for any account your staff held on the system named in the claim, check sender domains for lookalikes, and ask the company in writing what data about you that system held.

Is the payment deadline in a ransomware claim a real deadline?

It is a pressure device set by the group, not a legal or technical limit. It says nothing about whether the data is what the group describes, and paying does not buy reliable deletion: incident response firm Coveware documented in 2020 groups that re-extorted victims who had paid and posted data that had been paid for, and concluded that stolen data will not be credibly deleted.

Read next

Check any IP or domain

See what our indexed threat intelligence records say about it.

Check an indicator