# Condé Nast Data Breach: 32.8 Million User Records Offered for Sale Following WIRED Leak

> A database of 32,815,767 Condé Nast user records spanning Vogue, The New Yorker, GQ, Glamour, WIRED, Vanity Fair, and other publications has been listed for $15,000 on a Russian-language cybercrime forum. The dataset, described as the full collection behind the December 2025 WIRED leak, was verified by Ransomnews via a 5,000-record sample. Approximately 30.5 million records are not previously public. All records contain email addresses, 31.6% include full names, and 22.3% include postal addresses — no passwords, hashes, or payment-card data are present. Condé Nast has not publicly confirmed the breach.

- **Published:** 2026-09-07T00:00:00Z
- **Last reviewed:** 2026-09-07T00:00:00Z
- **Canonical:** https://intel.threadlinqs.com/threat/TL-2026-2383
- **Card image:** https://intel.threadlinqs.com/og/threat/threat-intel-conde-nast-user-data-sale-tl-2026-2383.webp?v=915d9954 (Threadlinqs threat intel: Condé Nast 32.8M Records for Sale: Detection & IOCs (TL-2026-2383) — High severity, 6 IOCs, 9 detection rules.)
- **ID:** TL-2026-2383
- **Severity:** HIGH
- **Category:** DATA_BREACH
- **Status:** ACTIVE
- **Actor:** Lovely
- **Detections:** 9 (1 shown free below) · **IOCs:** 6

## Description

On 7 September 2026, a listing appeared on a Russian-language cybercrime forum offering 32,815,767 Condé Nast user records for $15,000. The seller, a newly-registered account with little reputation offering escrow, claims the dataset is the full database behind the December 2025 WIRED leak. Ransomnews analyzed a 5,000-record sample and confirmed the data as genuine Condé Nast account information captured between September and late October 2025 through a series of IDOR and broken access control vulnerabilities.

The breach originated in Condé Nast's centralized identity platform — a shared account system used across all major publications including WIRED, Vogue, The New Yorker, GQ, Vanity Fair, Glamour, Allure, Architectural Digest, Bon Appétit, Condé Nast Traveler, SELF, Epicurious, Teen Vogue, Golf Digest, and Men's Journal. The platform's subscriber profiles were indexed by predictable, sequential user IDs, enabling an attacker to systematically iterate through these IDs and query backend APIs to retrieve profile data at scale. The backend APIs failed to consistently enforce authorization checks, and compoundingly, account management endpoints lacked sufficient authentication, reportedly allowing unauthenticated users to modify profile attributes such as email addresses and passwords. No API rate limiting or request throttling was in place to detect or block the bulk enumeration.

The attacker, using the alias 'Lovely,' initially contacted Dissent Doe of DataBreaches.net on 22 November 2025, posing as a security researcher seeking help with responsible disclosure. Lovely claimed to have found six vulnerabilities and reported they had only downloaded a small number of records as proof. After Condé Nast reportedly failed to respond for weeks, Lovely later admitted to having downloaded the entire database of over 33 million accounts. On 20 December 2025, Lovely leaked the WIRED subset of 2,366,576 records on the Breach Stars hacking forum, offering access for approximately $2.30 in forum credits and threatening to release 40+ million additional records.

The dataset for sale on the Russian-language forum contains 32,815,767 records, all with email addresses. First and last names appear in 31.6% of records, postal addresses in 22.3%, gender in 17.5%, date of birth in 12.6%, and phone numbers in 2.9%. A version of the dataset excluding WIRED contains 30,455,594 records — approximately 30.5 million records that have not surfaced publicly before. The sample's field completion rates matched the seller's claims within 1.2 percentage points. Validation checks showed 96.4% of U.S. ZIP codes matched the listed state, 93.5% matched the listed city, and 61.9% of full names had matching email addresses (versus 0.3% when shuffled). Account-creation dates in the sample range from February 1999 to 23 October 2025, with new accounts thinning sharply after September 2025 — consistent with extraction over several weeks during that period.

The data was independently verified by multiple parties. BleepingComputer validated 20 records as legitimate WIRED subscribers. Hudson Rock (co-founder Alon Gal) confirmed the authenticity of the WIRED leak by cross-referencing subscriber credentials against RedLine and Raccoon infostealer malware logs. The WIRED dataset was added to Have I Been Pwned on 27 December 2025. Troy Hunt noted that 81% of the records were already in HIBP from prior breaches.

While no passwords, password hashes, or payment-card data are present in the dataset, the exposed PII presents significant risks. A buyer could deploy targeted phishing campaigns referencing real subscriptions, names, and addresses to achieve high credibility. Physical mail scams referencing magazine titles are a risk given the 22.3% of records with postal addresses. The data can also be correlated with other breach datasets for identity fraud, used for credential-stuffing preparation, and leveraged for social engineering attacks. The absence of a public breach notification from Condé Nast raises potential regulatory exposure under GDPR (72-hour notification requirement), CCPA (California resident notification), and New York state breach notification laws. Condé Nast is privately held, so SEC disclosure rules do not apply; no regulatory fines or enforcement actions have been announced as of this writing.

## MITRE ATT&CK

- T1595 Active Scanning
- T1589 Gather Victim Identity Information
- T1190 Exploit Public-Facing Application
- T1133 External Remote Services
- T1059 Command and Scripting Interpreter
- T1106 Native API
- T1087 Account Discovery
- T1580 Cloud Infrastructure Discovery
- T1119 Automated Collection
- T1213 Data from Information Repositories
- T1048 Exfiltration Over Alternative Protocol
- T1567 Exfiltration Over Web Service
- T1565 Data Manipulation

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

**Bulk Profile Data Exfiltration via Identity API** (`TL-2026-2383-SPL-003`)

Detects a client that harvests an anomalous aggregate volume of profile payload bytes, or a request rate sustainable only by automation. Matches the collection and exfiltration leg of TL-2026-2383: scripted API calls (T1059/T1119) drained roughly 33 million subscriber records -- email, name, postal address, DOB, phone -- from the identity repository (T1213) over several weeks, with no rate limiting (CWE-770) to interrupt it, before the data was staged for sale (T1048/T1567).

- **Languages:** Splunk SPL, Microsoft KQL, Sigma
- **ATT&CK:** T1119, T1213, T1048, T1567, T1059
- **Data sources:** api_gateway_logs, network_egress_proxy, cdn_logs
- **Severity:** high
- **Confidence:** medium

### Splunk SPL

```spl
index=web sourcetype=api_access earliest=-30d latest=now
| regex uri_path='(?i)^/api/v[0-9]+/[^/]+/[0-9]+/(profile|details)'
| rex field=uri_path '(?i)/[^/]+/(?<object_id>[0-9]+)/(profile|details)'
| where isnum(object_id)
| stats count as requests, dc(object_id) as distinct_ids, sum(bytes) as total_bytes, avg(bytes) as avg_bytes, max(bytes) as max_bytes by src_ip
| eval exfil_mb = round(total_bytes / 1048576, 2)
| where exfil_mb > 50 OR (requests > 10000 AND avg_bytes > 20000)
| table src_ip, requests, distinct_ids, exfil_mb, avg_bytes, max_bytes
| sort - exfil_mb
```

### Microsoft KQL

```kql
// Bulk exfiltration of profile payloads: a client that pulls either a very
// large aggregate volume of profile bytes or a request rate only automation
// can sustain, matching the weeks-long harvesting of ~33M subscriber records.
let StartWindow = ago(24h);
let ProfilePattern = @"(?i)/[^/]+/([0-9]+)/(?:profile|details)";
W3CIISLog
| where TimeGenerated > StartWindow
| where csUriStem matches regex ProfilePattern
| extend objectId = tostring(extract_all(csUriStem, ProfilePattern)[0])
| summarize Requests = count(),
    DistinctIds = dcount(objectId),
    TotalBytes = sum(iff(isnull(csBytes), 0, csBytes)),
    AvgBytes = avg(iff(isnull(csBytes), 0, toreal(csBytes))),
    MaxBytes = max(iff(isnull(csBytes), 0, csBytes))
    by cIP
| extend ExfilMB = round(TotalBytes / 1048576.0, 2)
| where ExfilMB > 50 or (Requests > 10000 and AvgBytes > 20000)
| project cIP, Requests, DistinctIds, ExfilMB, AvgBytes, MaxBytes
| order by ExfilMB desc
```

### Sigma

```yaml
title: Sustained Bulk Download of User Profile Records
id: 9c1b4a7e-62f3-4d80-b5e9-1f0a72c8d3e4
status: experimental
description: >
  Detects a single source sustaining a very large number of profile-record
  retrievals over a short window. Matches the automated collection and
  exfiltration leg of TL-2026-2383, where roughly 33 million subscriber
  records were harvested from the Conde Nast identity platform across several
  weeks using scripted API calls (T1119, T1213) with no rate limiting in
  place (CWE-770) before being staged for sale (T1048, T1567).
references:
  - https://ransomnews.com/conde-nast-database-sale-2026/
  - https://haveibeenpwned.com/Breach/WIRED
author: Threadlinqs Team
date: 2026/09/08
tags:
  - attack.collection
  - attack.exfiltration
  - attack.t1119
  - attack.t1213
  - attack.t1048
logsource:
  category: webserver
  product: generic
detection:
  selection:
    cs-uri-stem|re: '/api/v[0-9]+/[^/]+/[0-9]+/(profile|details)'
  filter_known_agents:
    cs-user-agent|contains:
      - 'Googlebot'
      - 'bingbot'
      - 'pingdom'
      - 'uptimerobot'
  condition: (selection and not filter_known_agents) | count() by src_ip > 5000
falsepositives:
  - Major crawler networks (Googlebot, bingbot) with legitimate high traffic
  - Uptime or monitoring services (pingdom, uptimerobot) polling at high volume
  - Authorized mass exports of public profile data
level: high

```

**Known false positives:**

- Authorized data export or reporting jobs
- API response caching or CDN warming bursts
- Internal mirroring of profile data for analytics

**Tuning notes:** Threshold on aggregate bytes (default 50 MB); at roughly 1.5 KB per profile payload that represents about 34,000 records. Add a correlated egress-proxy rule for large uploads to public object storage or archive hosts (T1048/T1567). Exclude internal analytics exports and CDN warming, and extend the lookback window past 24h because the observed extraction ran for several weeks.

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

- Bulk Unauthenticated Profile Object Enumeration (IDOR) (locked: unlock with TL Creds)
- Bulk Email/Password Attribute Updates Across Distinct Accounts (locked: unlock with TL Creds)
- Automated Collection and Exfiltration of Subscriber Profiles (locked: unlock with TL Creds)
- Active Scanning and IDOR Enumeration of Profile API (locked: unlock with TL Creds)
- Mass Account-Data Modification to Distinct Accounts (locked: unlock with TL Creds)
- Sustained Bulk Download of User Profile Records (locked: unlock with TL Creds)
- Sequential User-ID Profile Enumeration (IDOR Exploitation) (locked: unlock with TL Creds)
- Mass Unauthorized Account Attribute Modification (Broken Access Control) (locked: unlock with TL Creds)

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

## Sources

- [Security Affairs — Condé Nast data of 32.8 million users offered for sale after WIRED leak](https://securityaffairs.com/198628/data-breach/conde-nast-data-of-32-8-million-users-offered-for-sale-after-wired-leak.html)
- [Ransomnews — Condé Nast database sale report (original verified sample analysis)](https://ransomnews.com/conde-nast-database-sale-2026/)
- [BleepingComputer — Hacker claims to leak WIRED database with 2.3 million records](https://www.bleepingcomputer.com/news/security/hacker-claims-to-leak-wired-database-with-23-million-records/)
- [SecurityWeek — Hacker Claims Theft of 40 Million Condé Nast Records After Wired Data Leak](https://www.securityweek.com/hacker-claims-theft-of-40-million-conde-nast-records-after-wired-data-leak/)
- [DataBreaches.net — Condé Nast gets hacked, and DataBreaches gets 'played'](https://databreaches.net/2025/12/25/conde-nast-gets-hacked-and-databreaches-gets-played-christmas-lump-of-coal-edition/)
- [Hudson Rock — WIRED Database Leaked: 40 Million Record Threat Looms for Condé Nast](https://www.hudsonrock.com/blog/wired-database-leaked-40-million-record-threat-looms-for-conde-nast)
- [Have I Been Pwned — WIRED Data Breach](https://haveibeenpwned.com/Breach/WIRED)
- [eSecurity Planet — 2.3M WIRED Subscriber Records Leaked in Condé Nast Data Breach (IDOR Technical Analysis)](https://www.esecurityplanet.com/threats/2-3m-wired-subscriber-records-leaked-in-conde-nast-data-breach/)
- [Cyber Security News — WIRED Database Breach: 2.3 Million Records Exposed](https://cybersecuritynews.com/wired-database-breach/)
- [SiliconANGLE — Millions of Wired user records leaked in claimed Condé Nast breach](https://siliconangle.com/2025/12/29/millions-wired-user-records-leaked-claimed-conde-nast-breach/)
- [HackRead — Hacker Behind Wired.com Leak Now Selling Full 40M Condé Nast Records](https://hackread.com/wired-com-hacker-data-leak-conde-nast-records/)
- [Infostealers.com — WIRED Database Leaked: 40 Million Record Threat Looms for Condé Nast](https://www.infostealers.com/article/wired-database-leaked-40-million-record-threat-looms-for-conde-nast/)
- [The Cyber Trove — WIRED Subscriber Data Breach: Condé Nast 2.3 Million Records](https://thecybertrove.com/wired-subscriber-data-breach-conde-nast-2-3-million-records/)
- [Dissent Doe (LinkedIn) — Does anyone have a connection at Condé Nast?](https://www.linkedin.com/posts/dissent-doe-phd-433281169_does-anyone-have-a-connection-at-cond%C3%A9-nast-activity-7404326441137369089-nmH_)
- [Troy Hunt (LinkedIn) — Have I Been Pwned: WIRED Data Breach](https://www.linkedin.com/posts/troyhunt_have-i-been-pwned-wired-data-breach-activity-7410836052115939328-ewto)

## 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-2026-2383
