AI is changing the scale and speed of data breach fallout. Criminals can now search stolen files for credentials, financial details and business relationships, extending the impact from one compromised company to the customers, suppliers and partners named in its records.
Lab 1’s 2026 study found that a breach exposed a median of 542 organizations beyond its initial victim, up from 482 the previous year. Across incidents from 2025 to 2026, the company identified 13,436 organizations exposed through other entities’ breaches. Those figures suggest that a breach’s reach increasingly depends on what attackers can extract from the data, not just which systems they initially compromise.
More data gives attackers more routes
Lab 1’s 2026 Anatomy of a Breach Report examined 328 million files across 2,089 breached datasets published online, including material stolen in ransomware extortion attacks. The median breach contained 29.8GB of data, 122% more than the 13.44GB median in 2025. The median file count rose 45%, from 22,647 to 32,734.
Lab 1 describes its source material as “ransomware and data-breach incidents in the public domain, reconstructed from forensic acquisitions of compromised systems.” Its analysis suggests that larger datasets offer criminals more opportunities for fraud, extortion and attacks that exploit details found in the files.
That risk is amplified by the way companies exchange information. Contracts, emails and supplier records can expose relationships across an entire network. Once attackers connect those details to credentials or personal information, they can pursue organizations that weren’t involved in the original breach.
Ricardo Amper, founder and CEO of identity verification vendor Incode Technologies, says the findings match what his company sees. “A stolen dataset used to sit on a forum until someone had the time to dig through it, but AI removed that bottleneck.”
“So every file now gets read, cross-referenced, and turned into a target list,” Amper says. “The blast radius keeps widening as the follow-on fraud lands on the suppliers, customers, and partners named inside the files.”
AI turns files into searchable intelligence
Lab 1 used AI agents to inspect the contents of the breached datasets, including unstructured files such as emails and PDFs. The analysis found that about 145 million files, or 44% of those studied, contained financial information such as banking details, tax identifiers and invoices.
Another 29 million files, or 9%, held HR material including payroll records, employment contracts and identity documents. Such information can help criminals make business email compromise attempts and other social engineering more convincing. Private cryptographic keys appeared in roughly 15% of incidents.
Lab 1 also identified seven threat actor groups conducting structured analysis of stolen data. The most advanced are using AI to index datasets and selling access to those capabilities, the company says. The first reported indexing-as-a-service offering appeared in March 2026.
According to Lab 1, the platform was designed to sort documents into categories criminals could monetize, map relationships, flag damaging files and estimate ransom demands. The intelligence available through such services reportedly included “share-buyback information for market manipulation, correspondence with sanctioned entities for use in ransomware negotiations, and security reports that could support follow-on attacks.”
One group, FulcrumSec, claimed it obtained an archive of 400,000 emails and singled out 61 files about work with US-sanctioned entities, which it could use to pressure the victim. The group also claimed it used AI agents to search 700,717 files and find 11,500 patient records associated with named clinical trials.
Exposure persists beyond the initial breach
“Historically, the sheer amount of time required to sift through breached datasets and the expertise needed to spot compromising information has held back the weaponization of data breaches,” says Robin Brattel, co-founder and CEO of Lab 1. “AI and criminal indexing services are removing those barriers.”
Brattel says exposed financial, HR and identity data can be searched and connected for fraud, attacks and extortion against organizations beyond the original victim. Zero Networks CEO Benny Lakunishok points to the risk of attackers using discovered information to gain access and move between systems or organizations.
“AI is making stolen data exponentially more valuable to criminals,” Lakunishok says. “What once took weeks to analyze can now be mined in minutes, revealing credentials, business relationships, and new targets”
What security leaders can do
For CISOs, the findings make third-party data exposure a direct concern. Organizations should know which suppliers hold sensitive information about them and have a clear path to rapid notification when a supplier is compromised.
Brattel distinguishes between secrets that can be replaced and records that can’t. “Credentials and keys can be revoked or rotated; invoices, contracts, supplier relationships, and internal communications cannot,” he tells CSO. “Once exposed, that information can remain useful for years from the attacker point of view.”
The practical takeaway is to treat information held by suppliers as part of the organization’s own exposure. A breach at a third party can leave a long trail of exploitable records, even after the original incident has been contained.