Another day, another claim of Dell data breach!

Hackers claim a second Dell data breach within a week, exposing sensitive internal files via compromised Atlassian tools. Allegedly, data from Jira, Jenkins, and Confluence was leaked. Dell is already investigating the first incident.

On September 19, 2024, _Hackread.com_ **published an exclusive report** detailing claims of a Dell data breach involving sensitive information related to 10,863 Dell employees. The same hacker behind the original breach is now claiming that Dell has been “breached again,” suggesting a larger ongoing issue.

The hacker, who operates under the alias “grep” on the infamous cybercrime platform Breach Forums, posted these claims on Sunday, September 22, 2024. In this post, “grep” revealed that Dell has suffered another significant breach, this time with the help of a fellow hacker known as “Chucky.” Together, they allegedly compromised Dell’s internal systems, exposing confidential data.

According to “grep,” the breach contains data related to Jira files, database tables, and schema migrations, amounting to a total of 3.5 GB of uncompressed data. The hackers claim to have gained access by compromising Dell’s Atlassian software suite, including Jenkins and Confluence, widely used tools for software development and collaboration.

Here’s the exact message shared by the hacker on the forum:

_Compromised data: Jira’s files, DB tables, schema migration, etc., totalling_ 3.5GB uncompressed. This time, it was breached by Chucky. Before Dell makes any claims, we both compromised your Atlassian and accessed Jenkins, Confluence, etc.  
GDPR said time is ticking, by the way, _xD._

The hacker on Breach Forum announcing his leak (Screenshot credit: Hackread.com)

In response to the first alleged breach, Dell told _Hackread.com_ that it was aware of the situation and had launched an investigation. However, as of now, the company has not addressed the latest breach claims.

The _Hackread.com_ research team has analyzed some of the files and determined that they likely contain sensitive information about Dell’s internal infrastructure. This includes system configurations, user credentials, security vulnerabilities, and development processes. If confirmed, this information could potentially be leveraged to target Dell’s systems on a much larger scale.

Further review of the leaked files suggests that they belong to a variety of enterprise tools and environments used by Dell, including Jira, database tables, schemas, and Atlassian tools such as Jenkins and Confluence. For security reasons, we are withholding specific details of the compromised files.

Screenshot from the leaked data Screenshot credit: Hackread.com)

_Hackread.com_ has reached out to Dell for a comment and will update this article as the story develops.

1.  **Hacker Leaks Data of 33,000 Accenture Employees**
2.  **Acer Online Store Hacked; 34,000 Customers’ Data Stolen**
3.  **Hacker Leaks Thousands of Microsoft and Nokia Employee Data**
4.  **Shadow IT: Personal GitHub Repos Expose Employee Cloud Secrets**
5.  **Nissan Confirms Data Breach Affected 100K Customers, Employees**

Hackers Claim Second Dell Data Breach in One Week

Apple’s macOS Sequoia update is causing major compatibility issues with popular security tools. Reportedly, users are facing disruptions…

Apple’s macOS Sequoia update is causing major compatibility issues with popular security tools. Reportedly, users are facing disruptions and frustration as vendors scramble to find solutions. Learn about the affected software, potential workarounds, and the latest updates on this ongoing issue.

Apple’s latest macOS update, macOS Sequoia (version 15), is reportedly causing issues with security tools from major providers like CrowdStrike, SentinelOne, ESET, and Microsoft. The update appears incompatible with several popular security tools, rendering them inoperable. The issue, which affects macOS users and enterprises, has caused frustration among those working with macOS-focused security tools. 

For your information, Apple officially launched its AI-focused macOS Sequoia on Monday. It is a new operating system, first revealed at WWDC 2024, offering exclusive features like Apple Intelligence, which uses Apple silicon to create language, images, and actions across apps, leveraging users’ personal context. 

However, the update has raised concerns for certain security-related products and applications as several security researchers and users on social media, Reddit, and a Mac-focused Slack channel reported issues with these security tools after installing Sequoia.

On Mastodon, security researcher Will Dormann reported firewall-related and DNS issues in macOS Sequoia, noting that blocking incoming connections within the macOS Sequoia firewall can also block replies to DNS requests, causing problems.

Dormann also noted a related issue affecting Chrome and Chromium-based browsers on macOS Sequoia, where blocking incoming connections for Google Chrome in the macOS firewall causes large downloads to stall.

Meanwhile, Apple has not yet addressed the concerns regarding Sequoia’s compatibility with security software such as ESET Endpoint Security and CrodStrike Falcon, and the full scope of the compatibility issues with Sequoia remains unclear.  

Most companies have advised users not to update the OS until the issue is resolved, as they are unable to support macOS Sequoia. CrowdStrike confirmed the issue and announced a delay in supporting Sequoia.

The company is waiting for Apple to release a fix before updating their software.  Similarly, ESET acknowledged network connection problems after the update but later clarified that their products are compatible with Sequoia. 

While SentinelOne initially advised users against upgrading to Sequoia, they later clarified that full support was available. However, some users still reported problems with other functionalities like firewalls and DNS configurations. 

A potential workaround shared by security researcher Wacław Jacek on his blog involves using command-line tools to adjust firewall settings for specific applications. According to Jacek, to modify firewall settings, use the CLI tool /usr/libexec/ApplicationFirewall/socketfilterfw.

This will allow your browser to access the internet again, but other software may still not function. To disable the entire firewall, open the terminal app, find the path to your web browser, run ls -l, and add your browser to the firewall using /usr/libexec/ApplicationFirewall/socketfilterfw.

Until the situation is resolved, please proceed with caution when considering the Sequoia update, especially if you rely heavily on third-party security software.

Mr. Mayuresh Dani, Manager, Security Research, at Qualys Threat Research Unit, shared his concern over the situation with Hackread.com stating, “With the release of new operating systems, all security vendors must test and qualify their releases. It is a good thing that security vendors have been proactive in this situation and have already sent out steps to take in case their systems are facing issues with the latest Mac update.”

Mayuresh highlighted that “From the looks of it, the networking stack – or the macOS Sequoia firewall to be specific – has changed because the security tools that use it to provide security are not able to do so. Not just security tools, VPNs are also having a difficult time getting a DNS resolution.”

He also suggested the following to the security teams responsible for securing Macs:

1.  Avoid updating to macOS Sequoia unless their security vendor has officially certified it for use.
2.  Turn off auto-updates to major OS releases before internal certification.
3.  Internally certify new operating system releases by installing Dev, and Beta builds of operating systems with certified software before organization-wide deployments.

Nevertheless, the importance of comprehensive testing before deploying major updates in production environments should not be ignored. It also highlights the need for organizations to have proper backup plans and alternative security measures in place. Developing a multi-layered security approach that doesn’t rely solely on a single tool or vendor can enhance overall stability.

1.  **CrowdStrike Update Causes Havoc By Disrupting Businesses**
2.  **Webroot Marked Facebook as Phishing Site, Windows as Malware**
3.  **Apple mistakenly approved malware hidden as Adobe Flash Player**
4.  **Microsoft Releases Tool to Fix CrowdStrike-Caused Windows Chaos**
5.  **Microsoft Defender Flags Tor Browser as Win32/Malgent!MTB Malware**

Apple’s macOS Sequoia Update Breaks Security Tools

A hacktivist group known as Twelve has been observed using an arsenal of publicly available tools to conduct destructive cyber attacks against Russian targets.
"Rather than demand a ransom for decrypting data, Twelve prefers to encrypt victims' data and then destroy their infrastructure with a wiper to prevent recovery," Kaspersky said in a Friday analysis.
"The approach is indicative of a

A hacktivist group known as Twelve has been observed using an arsenal of publicly available tools to conduct destructive cyber attacks against Russian targets.

"Rather than demand a ransom for decrypting data, Twelve prefers to encrypt victims' data and then destroy their infrastructure with a wiper to prevent recovery," Kaspersky said in a Friday analysis.

"The approach is indicative of a desire to cause maximum damage to target organizations without deriving direct financial benefit."

The hacking group, believed to have been formed in April 2023 following the onset of the Russo-Ukrainian war, has a track record of mounting cyber attacks that aim to cripple victim networks and disrupt business operations.

It has also been observed conducting hack-and-leak operations that exfiltrate sensitive information, which is then shared on its Telegram channel.

Kaspersky said Twelve shares infrastructural and tactical overlaps with a ransomware group called DARKSTAR (aka COMET or Shadow), raising the possibility that the two intrusion sets are likely related to one another or part of the same activity cluster.

"At the same time, whereas Twelve's actions are clearly hacktivist in nature, DARKSTAR sticks to the classic double extortion pattern," the Russian cybersecurity vendor said. "This variation of objectives within the syndicate underscores the complexity and diversity of modern cyberthreats."

The attack chains start with gaining initial access by abusing valid local or domain accounts, after which the Remote Desktop Protocol (RDP) is used to facilitate lateral movement. Some of these attacks are also carried out via the victim's contractors.

"To do this, they gained access to the contractor's infrastructure and then used its certificate to connect to its customer's VPN," Kaspersky noted. "Having obtained access to that, the adversary can connect to the customer's systems via the Remote Desktop Protocol (RDP) and then penetrate the customer's infrastructure."

Prominent among the other tools used by Twelve are Cobalt Strike, Mimikatz, Chisel, BloodHound, PowerView, adPEAS, CrackMapExec, Advanced IP Scanner, and PsExec for credential theft, discovery, network mapping, and privilege escalation. The malicious RDP connections to the system are tunneled through ngrok.

Also deployed are PHP web shells with capabilities to execute arbitrary commands, move files, or send emails. These programs, such as the WSO web shell, are readily available on GitHub.

In one incident investigated by Kaspersky, the threat actors are said to have exploited known security vulnerabilities (e.g., CVE-2021-21972 and CVE-2021-22005) in VMware vCenter to deliver a web shell that then was used to drop a backdoor dubbed FaceFish.

"To gain a foothold in the domain infrastructure, the adversary used PowerShell to add domain users and groups, and to modify ACLs (Access Control Lists) for Active Directory objects," it said. "To avoid detection, the attackers disguised their malware and tasks under the names of existing products or services."

Some of the names used include "Update Microsoft," "Yandex," "YandexUpdate," and "intel.exe."

The attacks are also characterized by the use of a PowerShell script ("Sophos\_kill\_local.ps1") to terminate processes related to Sophos security software on the compromised host.

The concluding stages entail using the Windows Task Scheduler to launch ransomware and wiper payloads, but not before gathering and exfiltrating sensitive information about their victims via a file-sharing service called DropMeFiles in the form of ZIP archives.

"The attackers used a version of the popular LockBit 3.0 ransomware, compiled from publicly available source code, to encrypt the data," Kaspersky researchers said. "Before starting work, the ransomware terminates processes that may interfere with the encryption of individual files."

The wiper, identical to the Shamoon malware, rewrites the master boot record (MBR) on connected drives and overwrites all file contents with randomly generated bytes, effectively preventing system recovery.

"The group sticks to a publicly available and familiar arsenal of malware tools, which suggests it makes none of its own," Kaspersky noted. "This makes it possible to detect and prevent Twelve's attacks in due time."

Found this article interesting? Follow us on Twitter __ and LinkedIn to read more exclusive content we post.

Hacktivist Group Twelve Targets Russian Entities with Destructive Cyber Attacks

The U.K. Information Commissioner's Office (ICO) has confirmed that professional social networking platform LinkedIn has suspended processing users' data in the country to train its artificial intelligence (AI) models.
"We are pleased that LinkedIn has reflected on the concerns we raised about its approach to training generative AI models with information relating to its U.K. users," Stephen

Privacy / Artificial Intelligence

The U.K. Information Commissioner's Office (ICO) has confirmed that professional social networking platform LinkedIn has suspended processing users' data in the country to train its artificial intelligence (AI) models.

"We are pleased that LinkedIn has reflected on the concerns we raised about its approach to training generative AI models with information relating to its U.K. users," Stephen Almond, executive director of regulatory risk, said.

"We welcome LinkedIn's confirmation that it has suspended such model training pending further engagement with the ICO."

Almond also said the ICO intends to closely keep an eye on companies that offer generative AI capabilities, including Microsoft and LinkedIn, to ensure that they have adequate safeguards in place and take steps to protect the information rights of U.K. users.

The development comes after the Microsoft-owned company admitted to training its own AI on users' data without seeking their explicit consent as part of an updated privacy policy that went into effect on September 18, 2024, 404 Media reported.

"At this time, we are not enabling training for generative AI on member data from the European Economic Area, Switzerland, and the United Kingdom, and will not provide the setting to members in those regions until further notice," Linked said.

The company also noted in a separate FAQ that it seeks to "minimize personal data in the data sets used to train the models, including by using privacy enhancing technologies to redact or remove personal data from the training dataset."

Users who reside outside Europe can opt out of the practice by heading to the "Data privacy" section in account settings and turning off the "Data for Generative AI Improvement" setting.

"Opting out means that LinkedIn and its affiliates won't use your personal data or content on LinkedIn to train models going forward, but does not affect training that has already taken place," LinkedIn noted.

LinkedIn's decision to quietly opt in all users for training its AI models comes only days after Meta acknowledged that it has scraped non-private user data for similar purposes going as far back as 2007. The social media company has since resumed training on U.K. users' data.

Last August, Zoom abandoned its plans to use customer content for AI model training after concerns were raised over how that data could be used in response to changes in the app's terms of service.

The latest development underscores the growing scrutiny of AI, specifically surrounding how individuals' data and content could be used to train large AI language models.

It also comes as the U.S. Federal Trade Commission (FTC) published a report that essentially said large social media and video streaming platforms have engaged in vast surveillance of users with lax privacy controls and inadequate safeguards for kids and teens.

The users' personal information is then often combined with data gleaned from artificial intelligence, tracking pixels, and third-party data brokers to create more complete consumer profiles before being monetized by selling it to other willing buyers.

"The companies collected and could indefinitely retain troves of data, including information from data brokers, and about both users and non-users of their platforms," the FTC said, adding their data collection, minimization, and retention practices were "woefully inadequate."

"Many companies engaged in broad data sharing that raises serious concerns regarding the adequacy of the companies' data handling controls and oversight. Some companies did not delete all user data in response to user deletion requests."

Found this article interesting? Follow us on Twitter __ and LinkedIn to read more exclusive content we post.

LinkedIn Halts AI Data Processing in UK Amid Privacy Concerns Raised by ICO

Plus: The FBI dismantles the largest-ever China-backed botnet, the DOJ charges two men with a $243 million crypto theft, Apple’s MacOS Sequoia breaks cybersecurity tools, and more.

The week was dominated by news that thousands of pagers, walkie-talkies and other devices were exploding across Lebanon on Tuesday and Wednesday in an attack targeting the militant group Hezbollah. At least 32 people were killed, including at least four children, and more than 3,200 people were injured. The covert campaign has widely been attributed to Israel, though none of the country's government agencies have commented.

In addition to the carnage, the attacks have—seemingly by design—had the effect of sowing paranoia and fear, not just among members of Hezbollah but also in the general Lebanese public. Hardware and warfare experts say that the incident is unlikely to establish a global precedent that people's most trusted communication devices and electronics, like smartphones, are rigged with explosives left and right. But it does create the potential to inspire copycats and puts defenders on notice that such attacks are possible.

Researchers say that China’s 2023 Zhujian Cup, a hacking competition with ties to the country's military, took the unusual step of requiring participants to keep the content of the exercise secret—and they may have been targeting a real victim as part of the event. Apple’s new stand-alone app Passwords that launched with iOS 18 may help solve your login problems. And a now-deleted post from billionaire Elon Musk that questioned why no one has attempted to assassinate Joe Biden and Kamala Harris renewed concerns this week that Musk is willing to inspire extremist violence and is a national security threat in the United States.

And there's more. Each week, we round up the privacy and security news we didn’t cover in depth ourselves. Click the headlines to read the full stories. And stay safe out there.

Last month, media outlets, Microsoft, and Google warned that an Iranian state-sponsored hacking group known as APT42 had targeted both the Joe Biden and Donald Trump political campaigns, and that it had successfully stolen emails from the Trump campaign that were later shared with reporters. Now the FBI has chimed in with the added revelation that the same hackers also sent those stolen Trump communications to the Democrats, too—though for now there's no sign that the Democrats solicited those emails from the Iranians or necessarily even received the Iranians' message.

Republicans were nonetheless quick to compare the news to accusations that the Trump campaign “colluded” with the Russian hackers, part of the Kremlin's GRU military intelligence agency, who breached the Democratic National Committee and the Clinton Campaign in 2016 to carry out a hack-and-leak operation. In a statement, the Trump campaign demanded that the Democrats “must come clean on whether they used the hacked material.” The Harris campaign told CNN that it has cooperated with law enforcement and that it was “not aware of any material being sent directly to the campaign,” believing the emails to be spam or phishing attempts. “We condemn in the strongest terms any effort by foreign actors to interfere in US elections, including this unwelcome and unacceptable malicious activity,” Morgan Finkelstein, the national security spokesperson for the Harris campaign, told CNN.

**FBI Dismantles the Largest-Ever Chinese State-Sponsored Botnet**

The FBI announced this week that it had taken down a network of hacked machines being secretly controlled by a Chinese state-sponsored hacking group known as Flax Typhoon. The botnet, made up of 260,000 routers and internet-of-things devices, was allegedly being run by a Chinese contractor known as the Beijing Integrity Technology Group, a rare instance of a known, publicly traded company operating essentially a massive collection of hacked devices on behalf of the Chinese state. The botnet, according to the FBI and security firm Black Lotus Labs, had been used to hack government agencies, defense contractors, telecoms, and other US and Taiwanese targets. At the time of its takedown, the botnet still encompassed 60,000 machines, making it the largest Chinese state-sponsored botnet ever, according to Black Lotus Labs.

**Two Men Charged With Stealing $243M in Cryptocurrency Using Social Engineering Scam**

On Wednesday night, two young men were arrested after they allegedly stole hundreds of millions of dollars of cryptocurrency and spent the earnings on luxury cars, watches, jewelry, and designer handbags. In an unsealed indictment, the US Department of Justice charged Malone Lam, 20, known online as “Anne Hathaway” and Jeandiel Serrano, 21, aka “VersaceGod,” with stealing $243 million in cryptocurrency and laundering the proceeds through mixing services to conceal the origin.

CoinDesk reported that the men allegedly tricked the heist’s victim, a creditor of the now-defunct trading firm Genesis, using a social engineering scam that led them to reset their Gemini two-factor authentication and transfer 4,100 bitcoin to a compromised wallet. An analysis of the transaction by blockchain investigator ZachXBT revealed that the $243 million was divided among multiple wallets and then distributed to over 15 exchanges.

**Apple MacOS Update Breaks Some Cybersecurity Tools**

On Thursday, TechCrunch reported that Apple's latest desktop operating system update, macOS 15 (Sequoia), breaks some functionality of major security tools made by CrowdStrike, SentinelOne, and Microsoft. It’s unclear what specifically in the update is causing the issues, but social media posts and internal Slack messages reviewed by the tech outlet show that the update has frustrated engineers working on macOS-focused security tools.

A CrowdStrike sales engineer informed colleagues via Slack, as seen by TechCrunch, that the company would not be able to support Sequoia on day one, despite its usual practice of quickly supporting new OS releases. While they hope for a quick patch, they will likely need to scramble to resolve the issue with an update in their own code, assuming no immediate fix is available from Apple, which has not yet commented on the issue.

**Leader of Crypto Extortion Gang Sentenced to 47 Years**

Cryptocurrency theft has become practically a common-garden form of cybercrime. But one brutal gang took that form of thievery to a new level of cruelty and violence, breaking into a series of victims' homes to threaten and extort them into handing over their crypto holdings, sometimes even resorting to kidnapping and torture. This week, that disturbing story came to a close with the sentencing of the group's ring leader, a Florida man named Remy St. Felix, to 47 years in prison. St. Felix is one of 12 members of the gang to have now been charged, convicted, and sentenced. Prior to the home invasions that St. Felix led, another member of the group named Jarod Seemungal allegedly stole millions with more traditional crypto hacking techniques. But St. Felix's more violent, offline extortion attempts netted his gang only around $150,000 in cryptocurrency before they were caught and sentenced to years behind bars. The lesson: Crime doesn't pay—or at least, not the physical kind.

Iranian Hackers Tried to Give Hacked Trump Campaign Emails to Dems

A North Korean advanced persistent threat (APT) actor (aka Gleaming Pisces) tried to sneak simple backdoors into public software packages.

Source: Chronicle via Alamy Stock Photo

One of North Korea's most sophisticated threat groups has been hiding remote access malware for macOS and Linux inside of open source Python packages.

North Korean advanced persistent threats (APTs) have become notorious for certain characteristic types of cyberattack in recent years. There's the cryptocurrency scam, which can come in many forms — often a fake trading platform, where victims are lured into divulging their wallet information or downloading malware. Supply chain attacks are common, particularly via poisoned packages typosquatting on public repositories. An impish recent trend involves contracting actual, honest labor to Western companies under false pretenses, then sending the salaries earned back to Kim's state. The reverse — agents posing as tech recruiters, convincing developers to download malware — is also common.

The group, which Palo Alto's Unit 42 tracks as Gleaming Pisces (and Microsoft as Citrine Sleet), seems to have supplemented category one with category two. Active since 2018, the financially motivated, DPRK Reconnaissance General Bureau (RGB)-linked group is known for attacks weaponizing fake crypto platforms. Unit 42 now assesses with medium confidence that it was responsible for uploading a handful of malicious packages to the Python Package Index (PyPI) back in February. The packages have since been taken down.

**DPRK-Poisoned PyPI Packages**

Most packages uploaded to open source repositories are simple by nature. As Louis Lang, co-founder and chief technology officer (CTO) at Phylum recalls, "What was interesting about these packages was that there was a higher order of complexity than you typically find among benign packages."

Phylum had identified four packages worth taking a second look at: real-ids, minisound, coloredtxt, and beautifultext. The innocuous names seemed to allude to legitimate functionality, like syntax highlighting for terminal outputs.

In reality, the packages contained malicious code that would be decoded and executed upon download. The code would then run bash commands in order to retrieve and download a remote access Trojan (RAT) called "PondRAT."

PondRAT is an entirely simple backdoor, capable of just a few functions: uploading and downloading files, checking to see that an implant is active or instructing it to sleep, and executing commands issued by the operator. It is, in essence, a "light" version of PoolRAT. PoolRAT is a known Gleaming Pisces backdoor for macOS that has a half dozen more standard capabilities than its successor, like listing directories, deleting files, etc.

**No Need for Windows**

More notable than the malware itself may be the fact that its authors wrote it only for macOS and Linux systems.

Forgoing hackers' long preferred Windows operating system makes sense, though, when one considers Gleaming Pisces' typical audience. As Lang explains, "They're targeting the actual builders, CI/CD infrastructure, developer workstations — environments that are overwhelmingly going to be Linux or macOS based. Very few people are doing development on straight Windows. So if you are targeting developers, it makes sense to ship variants for these systems, because that's where your target population lives."

Developers, then, need to be alert to phishing attacks, like those fake crypto platforms and job recruitment scams. Because while it's rare that anyone might pull an unpopular, ultra-generic package from PyPI, it's entirely likely that that same package could be quietly integrated into a broader infection chain.

"If you add a package, it could have downstream impacts, where you're actually pulling in 30, 40 other packages it may \[be connected to\]. So if I was a developer, I'd be very cognizant of what I'm installing, and try to minimize the attack surface by minimizing the number packages I'm pulling in. And then, obviously, scan the packages — look for these zombies, look for high-entropy strings, look for code obfuscation," Lang suggests.

"Like we always say," he adds, "you're one update away from malware."

**About the Author**

Nate Nelson is a freelance writer based in New York City. Formerly a reporter at Threatpost, he contributes to a number of cybersecurity blogs and podcasts. He writes "Malicious Life" -- an award-winning Top 20 tech podcast on Apple and Spotify -- and hosts every other episode, featuring interviews with leading voices in security. He also co-hosts "The Industrial Security Podcast," the most popular show in its field.

Citrine Sleet Poisons PyPI Packages With Mac &amp; Linux Malware

German authorities dismantled Boystown, a notorious Dark Web platform for CSAM, by deanonymizing Tor users in 2021. This…

German authorities dismantled Boystown, a notorious Dark Web platform for CSAM, by deanonymizing Tor users in 2021. This breakthrough raises concerns over Tor’s privacy as law enforcement targets criminal activities on the Dark Web.

German authorities, in collaboration with several international law enforcement agencies, successfully dismantled “Boystown,” a notorious dark web platform dedicated to child sexual abuse material (CSAM). This happened in April 2021 but the details of it have only been revealed now. What’s even crazier is that according to German media, authorities managed to deanonymize Tor users involved with the CSAM site and successfully arrest them.

With over 400,000 registered users, the site had been operational since 2019 and hosted some of the most severe forms of abuse, with many of the victims being young boys. The German Federal Criminal Police led this operation with support from Europol, alongside agencies from the Netherlands, the U.S., Canada, and several other countries.

Boystown’s administrators, three German men, were arrested, and a fourth suspect was detained in Paraguay, with extradition requested by Germany. These individuals helped users evade detection while facilitating the distribution of illegal content.

The platform was shut down in April 2021, and its chatrooms were also dismantled. This international operation marked a significant blow to illegal dark web activity involving CSAM.

****Deanonymizing Tor Nodes****

According to NDR’s report, German authorities’ ability to deanonymize users of Tor—a network designed for anonymity—played a crucial role in the success of this operation. Through surveillance of specific Tor nodes, they identified users accessing the site.

However, such breakthroughs have raised concerns about the safety of Tor’s anonymity features, even though the Tor Project insists its network remains secure. In its blog post, the Tor Project acknowledged being aware of the claims made in NDR’s report. However, the organization also stated that it had not received proof of concept (PoC) or documents independently verifying these claims from German authorities.

> _“_The Tor Project has not been granted access to supporting documents and has not been able to independently verify if this claim is true, if the attack took place, how it was carried out, and who was involved. In the absence of facts, it is hard for us to issue any official guidance or responsible disclosures to the Tor community, relay operators, and users at this time._“_
> 
> Isabela Fernandes – Executive Director of the Tor Project

Nevertheless, experts suggest that outdated software, combined with increased scrutiny of specific exit nodes, may have led to the identification of Boystown’s users. The implications of this case extend beyond just CSAM offenders. It has raised alarms among those who use Tor for legitimate purposes, such as whistleblowers and activists.

While privacy advocates fear that increased law enforcement surveillance of Tor could jeopardize users’ anonymity, authorities argue that targeting criminal activities like Boystown is a necessary step to protect the vulnerable.

This development indicates that law enforcement agencies may continue enhancing their surveillance of darknet activity. Though Tor remains a vital tool for privacy, these cases show that it is not entirely immune to infiltration, especially when outdated systems or weak links are involved.

If you are on the Tor browser stop illegal activities. For whistleblowers and journalists, here are five tips to help lay users better secure their Tor Browser:

1.  **Keep Tor Browser Updated**: Always use the latest version of the Tor Browser, as updates include important security patches that fix vulnerabilities. You can check for updates through the browser’s settings or download the newest version directly from the Tor Project website.
2.  **Disable JavaScript**: JavaScript can be used to exploit vulnerabilities and deanonymize users. To further enhance your security, disable JavaScript through the browser’s settings or use the “Safest” security level, which disables JavaScript by default. You can adjust the security settings by clicking the shield icon next to the address bar.
3.  **Avoid Installing Browser Add-ons**: Tor comes with privacy-enhancing features, and installing third-party add-ons could weaken your anonymity by introducing new vulnerabilities or leaking information unintentionally.
4.  **Use Bridges and Pluggable Transports**: If you are in a country or region that blocks Tor or monitors access to Tor nodes, use bridges or pluggable transports to circumvent censorship and avoid detection. These tools disguise your Tor traffic, making it harder for ISPs and governments to identify that you’re using Tor.
5.  **Don’t Use Personal Information**: Avoid logging into accounts that are linked to your real identity (e.g., personal email or social media) when using Tor. This undermines anonymity and could reveal your true identity through cookies or other tracking mechanisms. Consider using pseudonyms or temporary email accounts when browsing.

1.  **Tor Browser Flow Leaks Your Real IP Address**
2.  **23% of Tor browser relays found to be stealing Bitcoin**
3.  **What is Dark Web, Search Engines, What Not to Do on Dark Web**
4.  **Microsoft Defender Flags Tor Browser as Win32/Malgent!MTB Malware**
5.  **OnionPoison – Fake Tor Browser Installer Spreads Malware Via YouTube**

Police Broke Tor Anonymity to Arrest Dark Web Users in Major CSAM Bust

The company announced an update to its privacy policy, acknowledging it is using customer data to train its AI models.

Source: Chris Batson via Alamy Stock Photo

Professional social networking site LinkedIn allegedly used data from its users to train its artificial intelligence (AI) models, without alerting users it was doing so.

According to reports this week, LinkedIn hadn't refreshed its privacy policy to reflect the fact that it was harvesting user data for AI training purposes.

Blake Lawit, LinkedIn's senior vice president and general counsel, then posted on the company’s official blog that same day to announce that the company had corrected the oversight.

The updated policy, which includes a revised FAQ, confirms that contributions are automatically collected for AI training. According to the FAQ, LinkedIn's GenAI features could use personal data to make suggestions when posting.

**LinkedIn's AI Data-Gathering Is Automatic**

"When it comes to using members' data for generative AI training, we offer an opt-out setting," the LinkedIn post read. "Opting out means that LinkedIn and its affiliates won't use your personal data or content on LinkedIn to train models going forward, but does not affect training that has already taken place."

Shiva Nathan, founder and CEO of Onymos, expressed deep concern about LinkedIn's use of prior user data to train its AI models without clear consent or updates to its terms of service.

Related:Dark Reading Confidential: The CISO and the SEC

"Millions of LinkedIn users have been opted in by default, allowing their personal information to fuel AI systems," he said. "Why does this matter? Your data is personal and private. It fuels AI, but that shouldn’t come at the cost of your consent. When companies take liberties with our data, it creates a massive trust gap."

Nathan added this is not just happening with LinkedIn, pointing out many technologies and software services that individuals and enterprises use today are doing the same.

"We need to change the way we think about data collection and its use for activities like AI model training," he said. "We should not require our users or customers to give up their data in exchange for services or features, as this puts both them and us at risk."

LinkedIn did explain that users can review and delete their personal data from past sessions using the platform's data access tool, depending on the AI-powered feature involved.

**LinkedIn Faces Tricky Waters**

The US has no federal laws in place to govern data collection for AI use, and only a few states have passed laws on how users' privacy choices should be respected via opt-out mechanisms. But in other parts of the world, LinkedIn has had to put its GenAI training on ice.

Related:An AI-Driven Approach to Risk-Scoring Systems in Cybersecurity

"At this time, we are not enabling training for generative AI on member data from the European Economic Area, Switzerland, and the United Kingdom," the FAQ states, confirming that it has stopped the data collection in those geos.

Tarun Gangwani, principal product manager, DataGrail, says the recently enacted EU AI Act has provisions within the policy that require companies that trade in user-generated content be transparent about their use of it in AI modeling.

"The need for explicit permission for AI use on user data continues the EU's general stance on protecting the rights of citizens by requiring explicit opt-in consent to the use of tracking," Gangwani explains.

And indeed, the EU in particular has shown itself to be vigilant when it comes to privacy violations. Last year, LinkedIn parent company Microsoft had to pay out $425 million in fines for GDPR violations, while Facebook parent company Meta was slapped with a $275 million fine in 2022 for violating Europe's data privacy rules.

The UK's Information Commissioners Office (ICO) meanwhile released a statement today welcoming LinkedIn's confirmation that it has suspended such model training pending further engagement with the ICO.

"In order to get the most out of generative AI and the opportunities it brings, it is crucial that the public can trust that their privacy rights will be respected from the outset," ICO's executive director, regulatory risk, Stephen Almond said in a statement. "We are pleased that LinkedIn has reflected on the concerns we raised about its approach to training generative AI models with information relating to its UK users."

Related:How Shifts in Cyber Insurance Are Affecting the Security Landscape

Regardless of geography, it's worth noting that businesses have been warned against using customer data for the purposes of training GenAI models in the past. In August 2023, communications platform Zoom abandoned plans to use customer content for AI training after customers voiced concerns over how that data could be used. And in July, smart exercise bike startup Peloton was slapped with a lawsuit alleging the company improperly scraped data gathered from customer service chats to train AI models.

**About the Author**

Nathan Eddy is a freelance journalist and award-winning documentary filmmaker specializing in IT security, autonomous vehicle technology, customer experience technology, and architecture and urban planning. A graduate of Northwestern University’s Medill School of Journalism, Nathan currently lives in Berlin, Germany.

LinkedIn Addresses User Data Collection for AI Training

An Iranian advanced persistent threat (APT) threat actor likely affiliated with the Ministry of Intelligence and Security (MOIS) is now acting as an initial access facilitator that provides remote access to target networks.
Google-owned Mandiant is tracking the activity cluster under the moniker UNC1860, which it said shares similarities with intrusion sets tracked by Microsoft, Cisco Talos, and

An Iranian advanced persistent threat (APT) threat actor likely affiliated with the Ministry of Intelligence and Security (MOIS) is now acting as an initial access facilitator that provides remote access to target networks.

Google-owned Mandiant is tracking the activity cluster under the moniker **UNC1860**, which it said shares similarities with intrusion sets tracked by Microsoft, Cisco Talos, and Check Point as Storm-0861 (formerly DEV-0861), ShroudedSnooper, and Scarred Manticore, respectively.

"A key feature of UNC1860 is its collection of specialized tooling and passive backdoors that \[...\] supports several objectives, including its role as a probable initial access provider and its ability to gain persistent access to high-priority networks, such as those in the government and telecommunications space throughout the Middle East," the company said.

The group first came to light in July 2022 in connection with destructive cyber attacks targeting Albania with a ransomware strain called ROADSWEEP, the CHIMNEYSWEEP backdoor, and a ZEROCLEAR wiper variant (aka Cl Wiper), with subsequent intrusions in Albania and Israel leveraging new wipers dubbed No-Justice and BiBi (aka BABYWIPER).

Mandiant described UNC1860 as a "formidable threat actor" that maintains an arsenal of passive backdoors that are designed to obtain footholds into victim networks and set up long-term access without attracting attention.

Among the tools includes two GUI-operated malware controllers tracked as TEMPLEPLAY and VIROGREEN, which are said to provide other MOIS-associated threat actors with remote access to victim environments using remote desktop protocol (RDP).

Specifically, these controllers are designed to provide third-party operators an interface that offers instructions on the ways custom payloads could be deployed and post-exploitation activities such as internal scanning could be carried out within the target network.

Mandiant said it identified overlaps between UNC1860 and APT34 (aka Hazel Sandstorm, Helix Kitten, and OilRig) in that organizations compromised by the latter in 2019 and 2020 were previously infiltrated by UNC1860, and vice versa. Furthermore, both the clusters have been observed pivoting to Iraq-based targets, as recently highlighted by Check Point.

The attack chains involve leveraging initial access gained by opportunistic exploitation of vulnerable internet-facing servers to drop web shells and droppers like STAYSHANTE and SASHEYAWAY, with the latter leading to the execution of implants, such as TEMPLEDOOR, FACEFACE, and SPARKLOAD, that are embedded within it.

"VIROGREEN is a custom framework used to exploit vulnerable SharePoint servers with CVE-2019-0604," the researchers said, adding that it controls STAYSHANTE, along with a backdoor referred to as BASEWALK.

"The framework provides post-exploitation capabilities including \[...\] controlling post-exploitation payloads, backdoors (including the STAYSHANTE web shell and the BASEWALK backdoor) and tasking; controlling a compatible agent regardless of how the agent has been implanted; and executing commands and uploading/downloading files.

TEMPLEPLAY (internally named Client Http), for its part, serves as the .NET-based controller for TEMPLEDOOR. It supports backdoor instructions for executing commands via cmd.exe, upload/download files from and to the infected host, and proxy connection to a target server.

It's believed that the adversary has in its possession a diverse collection of passive tools and main-stage backdoors that align with its initial access, lateral movement, and information gathering goals.

Some of the other tools of note documented by Mandiant are listed below -

*   OATBOAT, a loader that loads and executes shellcode payloads
*   TOFUDRV, a malicious Windows driver that overlaps with WINTAPIX
*   TOFULOAD, a passive implant that employs undocumented Input/Output Control (IOCTL) commands for communication
*   TEMPLEDROP, a repurposed version of an Iranian antivirus software Windows file system filter driver named Sheed AV that's used to protect the files it deploys from modification
*   TEMPLELOCK, a .NET defense evasion utility that's capable of killing the Windows Event Log service
*   TUNNELBOI, a network controller capable of establishing a connection with a remote host and managing RDP connections

"As tensions continue to ebb and flow in the Middle East, we believe this actor's adeptness in gaining initial access to target environments represents a valuable asset for the Iranian cyber ecosystem that can be exploited to answer evolving objectives as needs shift," researchers Stav Shulman, Matan Mimran, Sarah Bock, and Mark Lechtik said.

The development comes as the U.S. government revealed Iranian threat actors' ongoing attempts to influence and undermine the upcoming U.S. elections by stealing non-public material from former President Donald Trump's campaign.

"Iranian malicious cyber actors in late June and early July sent unsolicited emails to individuals then associated with President Biden's campaign that contained an excerpt taken from stolen, non-public material from former President Trump's campaign as text in the emails," the government said.

"There is currently no information indicating those recipients replied. Furthermore, Iranian malicious cyber actors have continued their efforts since June to send stolen, non-public material associated with former President Trump's campaign to U.S. media organizations."

Iran's ramping up of its cyber operations against its perceived rivals also comes at a time when the country has become increasingly active in the Middle East region.

Late last month, the U.S. Cybersecurity and Infrastructure Security Agency (CISA) warned that the Iranian APT Lemon Sandstorm (aka Fox Kitten) has carried out ransomware attacks by clandestinely partnering with NoEscape, RansomHouse, and BlackCat (aka ALPHV) groups.

Censys' analysis of the hacking group's attack infrastructure has since uncovered other, currently active hosts that are likely part of it based on commonalities based on geolocation, Autonomous System Numbers (ASNs), and identical patterns of ports and digital certificates.

"Despite attempts at obfuscation, diversion, and randomness, humans still must instantiate, operate, and decommission digital infrastructure," Censys' Matt Lembright said.

"Those humans, even if they rely upon technology to create randomization, almost always will follow some sort of pattern whether it be similar Autonomous Systems, geolocations, hosting providers, software, port distributions or certificate characteristics."

Found this article interesting? Follow us on Twitter __ and LinkedIn to read more exclusive content we post.

Iranian APT UNC1860 Linked to MOIS Facilitates Cyber Intrusions in Middle East

How the Kimsuky nation-state group and other threat actors are exploiting poor email security — and what organizations can do to defend themselves.

Dr. Sean Costigan, Managing Director, Resilience Strategy, Red Sift

September 20, 2024

5 Min Read

Source: Brian Jackson via Alamy Stock Photo

COMMENTARY

With heightened geopolitical tensions, a surge in cyberattacks on US and allied organizations by a North Korean cyber-espionage group is hardly unexpected. What is disquieting, however, is that an advanced persistent threat (APT) group known as Kimsuky has seen remarkable success by turning a defensive strength into a weakness — exploiting poorly configured Domain-based Message Authentication, Reporting and Conformance (DMARC) policies to carry out spear-phishing campaigns to secure advantage.

A May 2 advisory from the FBI, the National Security Agency (NSA), and the US State Department stated that Kimsuky, acting as an arm of North Korea's Reconnaissance General Bureau (RGB), has been sending spoofed emails to individuals in high-profile think tanks, media outlets, nonprofits, academia, and other organizations. The emails are part of an intelligence campaign to troll for information on geopolitics and foreign policy plans, particularly related to nuclear policies, sanctions, and other sensitive concerns involving the Korean peninsula.    

With sanctions biting, North Korea has developed a formidable cybercrime capability to generate liquidity for the regime. However, in this case, we see Kimsuky threat actors alter their focus to intelligence operations, targeting troves of information held by trusted parties and prominent organizations. Although the ongoing campaign has complex geopolitical implications, effectively defending against these attacks fundamentally relies on robust, actionable, and properly executed cyber-hygiene practices.

Related:Singapore Arrests 6 Suspected Members of African Cybercrime Group

**DMARC Misconfigurations Are Too Common**

Kimsuky is using trusted networks with improperly configured or missing DMARC to spoof legitimate domains and impersonate trusted personalities and organizations. The DMARC protocol was created to stop the compromise of user accounts and hinder the very types of social engineering at work here.

This is how it's supposed to work: DMARC allows email recipients to verify an email's origin through the Domain Name System (DNS), ensuring that threat actors cannot spoof legitimate domains. DMARC checks the Sender Policy Framework (SPF) and DomainKeys Identified Mail (DKIM) records for an incoming email and, if it does not appear to be legitimate, tells the receiving email server what to do next.

But as Kimsuky's attacks have shown, that only works if DMARC services are properly configured. As the IC3 advisories detail, misconfigurations are far too common or policies are poorly defined by the domain owners. For some organizations, self-managing DMARC may seem cost-effective, but it can also lead to significant oversights, including increasing vulnerabilities, failing to pay heed to evolving threats, missing sound compliance reporting, and creating a false sense of security.  

Related:Indian Army Propaganda Spread by 1.4K AI-Powered Social Media Accounts

**What North Korea's Attack Looks Like**

Kimsuky's spear-phishing campaigns may begin with an innocuous email from a seemingly credible source, building trust before sending a subsequent email with a malicious link or attachment. The group then uses successful compromises to escalate attacks with more credible spear-phishing emails aimed at higher-value targets.

The group focuses its intelligence-gathering activities against South Korea, Japan, and the United States, targeting individuals identified as experts in various fields. According to a subsequent advisory from the Cybersecurity and Infrastructure Security Agency (CISA), think tanks and South Korean government entities have also been targeted.  

One real-world example from the FBI-NSA advisory had a subject line reading: "\[Invitation\] US Policy Toward North Korea Conference." The message, seemingly from a known university, begins: "I hope you and your family are enjoying a lovely holiday and a restful season. It is my privilege to invite you to provide a keynote address for a private workshop, hosted by the \[legitimate think tank\] to discuss the U.S. policy toward North Korea." As further inducement, the email also offers a $500 speaker's fee.

Related:Chinese Threat Actors Use MSI Files to Bypass Windows, VT Detection

Another email had the subject line "Questions about N. Korea," with the writer posing as a journalist from a legitimate media outlet and requesting an interview, followed by a broad outline of North Korea's nuclear activities.

In the university example, the email received a "pass" from SPF and DKIM checks, suggesting the attacker gained access to the university's legitimate email client. And although DMARC returned a "fail" because the sender's email domain differed from SPF and DKIM records for the legitimate source, the organization's DMARC policy was not set to take filtering action, so the message was delivered. In the second case, no DMARC policy was present, allowing the attacker to spoof the journalist's name and the news organization's email domain.

**Why DMARC Matters**

The US government's advisories offer compelling reasons for organizations to secure their digital estates. Kimsuky is not alone among APTs nor, more broadly, cybercriminals who work for profit: Lessons are shared and all are becoming increasingly savvy at targeting misconfigurations and weaknesses.

Securing and properly configuring DMARC is key since it improves organizational cyber hygiene and broadly protects against ubiquitous threats like business email compromise and ransomware email attacks.

Notably, industry or regulatory requirements may already make DMARC a requirement for your organization. As of February 2024, Google and Yahoo have required DMARC for organizations sending large volumes of email, and Microsoft is reportedly planning to follow suit. Additionally, the PCI DSS 4.0 requires implementation of DMARC. According to BIMI Radar, since the FBI's May 2 advisory, DMARC adoption globally has grown from 3.74 million organizations to 5.71 million organizations, as of June 17. 

There's a business imperative at work as well. Organizations must prioritize cyber hygiene to safeguard their digital assets, prevent data breaches, and protect against evolving cybersecurity threats. DMARC should be part of your organization's cyber posture. When properly managed, not only does it ensure better deliverability, provide protection against phishing and business email compromise (BEC), and enable the deployment of Brand Indicators for Message Identification (BIMI), but it can help close doors against nation-state espionage and cybercrime.

**About the Author**

Managing Director, Resilience Strategy, Red Sift

Sean Costigan is an expert in emerging security challenges and a highly sought-after speaker on technology and national security. He is the lead for NATO’s cybersecurity curriculum and is widely published on national security matters relating to information security and hybrid threats. He is also a Professor at the George C. Marshall Center, where he educates on global cybersecurity, hybrid warfare, crime, and national security.

Tag

#microsoft