Researchers at Bitglass have tracked the journey of a series of fake accounts “leaked” to the Dark Web to have an overview of login attempts and how such data travels across the underground.
This is the second operation of its kind conducted by Bitglass, after the company tracked fake personal data across the globe in a similar experiment last year. The company discovered at the time that the leaked data traveled five continents and 22 countries within two weeks and that it was viewed more than 1,000 times and downloaded 47 times.
Dubbed Project Cumulus, the second annual “Where’s Your Data” experiment (PDF) involved creating a digital identity for an employee of a fictitious retail bank, along with a functional web portal for the bank, and a Google Drive account, and pairing them with real credit-card data. Researchers leaked “phished” Google Apps credentials to the Dark Web and started tracking the activity on these accounts.
Bitglass researchers registered five bank login attempts and three Google Drive login attempts within 24 hours after the so called leak. Within 48 hours of the initial leak, files were downloaded, and the account was viewed hundreds of times over the course of a month, with many hackers successfully accessing the victim’s other online accounts.
The researchers recorded more than 1400 visits to the Dark Web credentials and the fictitious bank’s web portal and note that one in ten hackers attempted to log in to Google with the leaked credentials. Moreover, they say that 94 percent of hackers who accessed the Google Drive uncovered the victim’s other online accounts and attempted to log into the bank’s web portal.
According to the researchers, 36 percent of hackers successfully accessed the fake victim’s personal banking account using the leaked password, while some of them logged into the account several times. Bitglass also noticed that 68 percent of all logins came from Tor-anonymized IP addresses to hide their tracks.
Last year, when the leaked files were viewed 200 times in the first few days, only few download attempts over Tor were recorded, making files easy to track. However, the researchers noticed a large number of downloads via Tor late last year, after an eight-month quiet period.
The experiment also revealed that 12 percent of hackers who successfully accessed the Google Drive attempted to download files with sensitive content and that some of them even cracked encrypted files after download. Researchers were able to track this type of activity because all files were embedded with Bitglass watermarks that messaged back each time new activity was recorded.
The researchers also note that 34.85 percent of the non-Tor visits to the fake bank web portal were from Russia, 15.67 percent came from the United States, 3.5 percent were from China, and 2 percent came from Japan. Overall, login attempts were registered from 30 countries across six continents, the researchers said.
“Our second data-tracking experiment reveals the dangers of reusing passwords and shows just how quickly phished credentials can spread, exposing sensitive corporate and personal data. Organizations need a comprehensive solution that provides a more secure means of authenticating users and enables IT to quickly identify breaches and control access to sensitive data,” Nat Kausik, CEO, Bitglass, said.
Related: DoJ Investigating Leak of FBI, DHS Employee Details
Related: 13 Million Passwords Leaked From Free Hosting Service

More from SecurityWeek News
- Threat Hunting Summit Virtual Event NOW LIVE
- Video: ESG – CISO’s Guide to an Emerging Risk Cornerstone
- Threat Modeling Firm IriusRisk Raises $29 Million
- SentinelOne Announces $100 Million Venture Fund
- Today: 2022 CISO Forum Virtual Event
- Cymulate Closes $70M Series D Funding Round
- SecurityWeek to Host CISO Forum Virtually September 13-14, 2022: Registration is Open
- Privilege Escalation Flaw Haunts VMware Tools
Latest News
- In Other News: AI Regulation, Layoffs, US Aerospace Attacks, Post-Quantum Encryption
- Blackpoint Raises $190 Million to Help MSPs Combat Cyber Threats
- Google Introduces SAIF, a Framework for Secure AI Development and Use
- ‘Asylum Ambuscade’ Group Hit Thousands in Cybercrime, Espionage Campaigns
- Evidence Suggests Ransomware Group Knew About MOVEit Zero-Day Since 2021
- SaaS Ransomware Attack Hit Sharepoint Online Without Using a Compromised Endpoint
- Google Cloud Now Offering $1 Million Cryptomining Protection
- Democrats and Republicans Are Skeptical of US Spying Practices, an AP-NORC Poll Finds
