# N0n ransomware group claims data theft from MediaMath (US, Technology) — 143M-row identity graph

> On 2026-11-16 the N0n data-theft extortion group listed MediaMath, a US programmatic advertising and identity-resolution firm, on its leak site with status 'NOT PAID — EXPOSED'. The group claims a 143,186,369-row cookie-to-device identity graph plus campaign analytics and client data; the claims are unverified and MediaMath has not confirmed a breach.

- **Published:** 2026-11-16T13:04:32Z
- **Last reviewed:** 2026-11-16T13:04:32Z
- **Canonical:** https://intel.threadlinqs.com/threat/TL-2026-6143
- **Card image:** https://intel.threadlinqs.com/og/threat/TL-2026-6143.png
- **ID:** TL-2026-6143
- **Severity:** HIGH
- **Category:** RANSOMWARE
- **Status:** ACTIVE
- **Detections:** 9 (1 shown free below) · **IOCs:** 12 (3 shown free below)

## Description

ransomware.live recorded MediaMath (United States, Technology) as a victim of the N0n group, discovered 2026-11-16 12:42 UTC, with an estimated attack date of 2026-11-16 and status 'NOT PAID — EXPOSED'. According to the listing, the group claims to hold: a consumer identity graph export of 143,186,369 rows linking consumer cookie identifiers to device identifiers; campaign analytics exports containing agency and advertiser names, budgets and performance for well-known consumer brands; curated marketplace consumer segment memberships; brand safety and fraud verification data; a named client's site whitelist; and monthly campaign performance and verification reports. No ransom amount is listed. All details are the group's own claims, relayed by a single aggregator; no victim statement or independent verification was found. No CVE, IOC or intrusion vector is stated for this specific victim.

N0n (also written n0n) is a data-theft extortion group that emerged in September 2026. ransomware.live tracks 281 N0n victims (159 posted in the current month), 49% in the United States, and reports that 78.9% of victims with a domain are tied to infostealer activity; many entries are marked 'PENDING EXPOSURE'. WatchGuard classifies the group as a 'Data Broker' running direct and double extortion from a single Tor leak site. On 2026-10-12 the group posted 38 maritime, port and logistics organizations in one batch with a 2026-10-19 deadline; reported demands ranged from roughly $2.4M to $28M depending on the source, and none of the 38 had publicly confirmed a breach.

SOCRadar (published 2026-10-30) documented the operator's exposed workspace — a public Python HTTP file server at 185.157.161.148:8877 (Sweden) holding 1,719 files including attack scripts, ransomware source code, extortion infrastructure and an AI coding agent environment (OpenCode, 154 model entries across seven providers). Observed access methods were mass RDP scanning, automated FortiGate SSL-VPN credential stuffing (GET /remote/login) through a 15-thread residential SOCKS5 proxy pool, exposed cloud (AWS) keys and reused hosting-portal sessions; SOCRadar found no evidence of vulnerability exploitation. Post-access activity included hidden administrator accounts disguised as services, covert proxies, credential/session-token theft, mailbox access with captured email MFA codes, S3 data exfiltration (40 buckets, up to 110 GB each), encryption, and deletion of data and backups. The locker used hardcoded, reused encryption keys. SOCRadar suspects a Russia-based operator but found no definitive attribution. Whether any of these TTPs were used against MediaMath is not stated in any source; they are group-level behaviours and should be treated as hunting guidance, not confirmed MediaMath facts.

## MITRE ATT&CK

- T1595 Active Scanning
- T1588.007 Obtain Capabilities: Artificial Intelligence
- T1133 External Remote Services
- T1078.004 Valid Accounts: Cloud Accounts
- T1110.004 Brute Force: Credential Stuffing
- T1539 Steal Web Session Cookie
- T1136 Create Account
- T1021.001 Remote Services: Remote Desktop Protocol
- T1090 Proxy
- T1114 Email Collection
- T1530 Data from Cloud Storage
- T1567 Exfiltration Over Web Service
- T1490 Inhibit System Recovery
- T1485 Data Destruction
- T1657 Financial Theft

## Free detection (Splunk SPL, Microsoft KQL, Sigma)

**N0n FortiGate SSL-VPN Credential Stuffing (/remote/login) — Initial Access** (`TL-2026-6143-SPL-001`)

Detects bursts of failed FortiGate SSL-VPN logins in which one source address tries many distinct usernames within 10 minutes. This is the pattern of N0n's automated credential-stuffing pipeline against GET /remote/login (T1110.004), which SOCRadar reports runs through a 15-thread residential SOCKS5 proxy pool. The pipeline feeds valid-account access through External Remote Services (T1133). TL-2026-6143 / MediaMath is an unverified leak-site claim; this is group-level hunting content, not a confirmed MediaMath TTP. No CVE is involved: access relied on weak or stolen credentials.

- **Languages:** Splunk SPL, Microsoft KQL, Sigma
- **ATT&CK:** T1110.004, T1133, T1595
- **Data sources:** fortinet:fortigate_event (index=fortinet, sourcetype=fortigate_event), CommonSecurityLog (Fortinet CEF)
- **Severity:** high
- **Confidence:** high

### Splunk SPL

```spl
index=fortinet sourcetype=fortigate_event subtype=vpn earliest=-60m (action="ssl-login-fail" OR logdesc="SSL VPN login fail" OR logid=0101039426)
| eval src_ip=coalesce(remip, srcip)
| bin _time span=10m
| stats count AS failed_logins dc(user) AS distinct_users values(user) AS sample_users BY _time src_ip
| where failed_logins>=20 AND distinct_users>=5
| sort - failed_logins
```

### Microsoft KQL

```kql
CommonSecurityLog
| where TimeGenerated > ago(1h)
| where DeviceVendor =~ 'Fortinet'
| where Activity has 'SSL VPN login fail' or DeviceEventClassID == '0101039426' or AdditionalExtensions has 'ssl-login-fail'
| extend SourceAddr = coalesce(SourceIP, tostring(extract(@'remip=([0-9.]+)', 1, AdditionalExtensions)))
| extend AcctName = coalesce(SourceUserName, tostring(extract(@'user=([^; ]+)', 1, AdditionalExtensions)))
| summarize FailedLogins = count(), DistinctUsers = dcount(AcctName), SampleUsers = make_set(AcctName, 10) by SourceAddr, bin(TimeGenerated, 10m)
| where FailedLogins >= 20 and DistinctUsers >= 5
| project TimeGenerated, SourceAddr, FailedLogins, DistinctUsers, SampleUsers
```

### Sigma

```yaml
title: FortiGate SSL-VPN Failed Login (N0n Credential Stuffing Building Block)
id: a3e91c52-6f0b-4d7e-8c24-91b5d7e03a11
status: experimental
description: Detects FortiGate SSL-VPN login failures. N0n (TL-2026-6143) runs automated credential stuffing against /remote/login through a residential SOCKS5 proxy pool. A single event is weak signal; alert on 20+ events from one source or 5+ distinct usernames within 10 minutes using the SIEM's aggregation or correlation feature.
references:
  - https://socradar.io/blog/n0n-ransomware-ai-assisted-extortion/
  - https://www.ransomware.live/group/n0n
author: Threadlinqs Team
date: 2026-11-16
tags:
  - attack.credential-access
  - attack.t1110.004
  - attack.initial-access
  - attack.t1133
logsource:
  product: fortinet
  service: fortigate
detection:
  selection:
    logdesc: 'SSL VPN login fail'
  condition: selection
falsepositives:
  - Users mistyping passwords or locked-out accounts after a password change
  - Misconfigured VPN clients retrying stale credentials
level: medium
```

**Known false positives:**

- Legitimate users repeatedly failing VPN login after a password rotation or account lockout
- Vulnerability scanners or authorized red-team credential tests against the SSL-VPN portal
- Shared NAT egress addresses (branch offices, mobile carriers) where many real users fail logins at once

**Tuning notes:** Adjust index and sourcetype to your Fortinet ingestion. Raise failed_logins/distinct_users thresholds for large user populations, and exclude corporate NAT egress ranges. N0n rotates residential proxy IPs, so also hunt for one username failing from many IPs, and correlate any successful ssl-login after a failure burst. Enforce phishing-resistant MFA and lockout on the portal.

### 8 more detection rules (Splunk SPL, Microsoft KQL, Sigma), unlock with TL Creds

- N0n External RDP Failed-Logon Spray Followed by Success — Initial Access (locked: unlock with TL Creds)
- N0n Bulk S3 Read/Delete via IAM User Access Key — Cloud Collection and Destruction (locked: unlock with TL Creds)
- N0n Covert Proxy and Tunneling Tool Execution — C2 Infrastructure (locked: unlock with TL Creds)
- N0n Exposed AWS Key Use: S3 ListBuckets Enumeration by IAM User — Initial Access (locked: unlock with TL Creds)
- N0n Backup and Shadow Copy Destruction Before Encryption — Impact (locked: unlock with TL Creds)
- N0n Tor Leak-Site and .onion Resolution from Corporate Host — C2 / Extortion Contact (locked: unlock with TL Creds)
- N0n Hidden Service-Named Local Admin Account Creation — Post-Exploitation Persistence (locked: unlock with TL Creds)
- N0n Operator Infrastructure Egress ([indicator]:8877) — C2 / Tool Staging (locked: unlock with TL Creds)

Sign up free to unlock the detection rules: https://threadlinqs.com/?modal=signup

## Sample indicators (3 of 12)

Defanged, TLP:CLEAR.

- IP address: `185.157.161[.]148`
- Domain: `nongzecboljwv3yfndkggsybsglfrkffw7bvk2zemuteoxe6etpusnad[.]onion`
- URL: `hxxps://www[.]ransomware[.]live/id/TWVkaWFNYXRoQE4wbg`

Full set of 12 indicators unlocks with TL Creds: https://threadlinqs.com/tl-creds

## Sources

- [ransomware.live victim entry: MediaMath (N0n)](https://www.ransomware.live/id/TWVkaWFNYXRoQE4wbg)
- [ransomware.live group profile: N0n](https://www.ransomware.live/group/n0n)
- [SOCRadar: n0n ransomware AI-assisted extortion (exposed operator server)](https://socradar.io/blog/n0n-ransomware-ai-assisted-extortion/)
- [SOCRadar: n0n group global ports and shipping leak](https://socradar.io/blog/n0n-group-global-ports-shipping-leak/)
- [The Register: n0n ransomware coverage](https://www.theregister.com/a/5306754)
- [Tech Insider: n0n group ransomware maritime firms 2026](https://tech-insider.org/n0n-group-ransomware-maritime-firms-2026/)
- [WatchGuard Ransomware Tracker: n0n](https://www.watchguard.com/it/wgrd-security-hub/ransomware-tracker/n0n)
- [Crypto Briefing: N0n ransomware group spends $7K on AI tools, then leaves its own server wide open](https://cryptobriefing.com/n0n-ransomware-ai-tools-exposed-server/)

## Full data

The other 8 detection rules (Splunk SPL / Microsoft KQL / Sigma) unlock with TL Creds; a free account includes daily creds: https://threadlinqs.com/?modal=signup
Indicator values unlock with TL Creds, at a lower price on Red and Purple; AI agents can query rules and IOCs through the Threadlinqs MCP server (Purple): https://intel.threadlinqs.com/mcp

Canonical: https://intel.threadlinqs.com/threat/TL-2026-6143
