Pierluigi Paganini
June 19, 2025
Researchers announced the discovery of what appears to be the largest data breach ever recorded, with an astonishing 16 billion login credentials exposed. The ongoing investigation, which began earlier this year, suggests that the credentials were collected through multiple infostealer malware strains.
Cybernews researchers who discovered the data leak, reported that tit was composed of 30 massive leaked datasets across various platforms, totaling an unprecedented 16 billion exposed login records.
“Our team has been closely monitoring the web since the beginning of the year. So far, they’ve discovered 30 exposed datasets containing from tens of millions to over 3.5 billion records each.” reads the report published by CyberNews. “In total, the researchers uncovered an unimaginable 16 billion records.”
The researchers state that, except for one previously reported case, all 30 leaked datasets are newly discovered. Alarming new leaks keeps surfacing, showing infostealer malware is widespread.
However, the exposed datasets were only briefly accessible, the experts state that most of them were exposed on unsecured Elasticsearch or storage instances.
“This is not just a leak – it’s a blueprint for mass exploitation. With over 16 billion login records exposed, cybercriminals now have unprecedented access to personal credentials that can be used for account takeover, identity theft, and highly targeted phishing. What’s especially concerning is the structure and recency of these datasets – these aren’t just old breaches being recycled. This is fresh, weaponizable intelligence at scale,” researchers said.
CyberNews researchers speculate that most of the 16 billion leaked records came from stealer malware, credential stuffing, and old breaches. The data, structured by URL, login, and password, targets services like Apple, Google, Facebook, Telegram, GitHub, and even government portals. While many records overlap, the true number of exposed accounts is still unclear. The datasets, briefly exposed via unsecured servers, now pose major risks, fueling phishing, ransomware, and account takeovers.
The leaked datasets range from 16 million to 3.5 billion records, averaging 550 million each. Some datasets had generic names like “logins,” while others hinted at origins, such as Telegram or Russia.
The data likely gathered by infostealer often includes tokens, cookies, and sensitive metadata.
It is my personal opinion that we are not facing ‘the mother of all breaches’: the 16 billion passwords do not come from a new attack on online companies, but from old stolen and aggregated credentials. This is a collection of data that has already been compromised through infostealers and phishing campaigns, often belonging to datasets that have been circulating for years. We should analyze the composition of these datasets in detail before making any serious judgment. It’s important to highlight the need to use unique passwords, enable MFA, and always use up-to-date anti-malware systems, as well as keep the software on our machines updated.
In the summer of 2024, CyberNews discovered the largest password compilation to date, known as RockYou2024, on a popular hacking forum.
The compilation (“rockyou2024.txt”) contains 9,948,575,739 unique plaintext passwords was posted on July 4th, 2024, by a user with the handle “ObamaCare.” The experts believe the collection contains data from “old and new data breaches.”
The RockYou2024 compilation is an expansion of the RockYou2021 collection that was discovered in 2021.
RockYou2021 had 8.4 billion entries of passwords, which have presumably been combined from previous data leaks and breaches. The compilation has been named ‘RockYou2021’ by the forum user, presumably in reference to the RockYou data breach that occurred in 2009, when threat actors hacked their way into the social app website’s servers and got their hands on more than 32 million user passwords stored in plain text.
Follow me on Twitter: @securityaffairs and Facebook and Mastodon
(SecurityAffairs – hacking, data breach)
