Darktrace cyber analysts are world-class experts in threat intelligence, threat hunting and incident response, and provide 24/7 SOC support to thousands of Darktrace customers around the globe. Inside the SOC is exclusively authored by these experts, providing analysis of cyber incidents and threat trends, based on real-world experience in the field.
Written by
Max Heinemeyer
Global Field CISO
Share
06
Jan 2021
For a high-level explanation of the SolarWinds hack, watch our video below.
The SUNBURST malware attacks against SolarWinds have heightened companies’ concerns about the risk to their digital environments. Malware installed during software updates in March 2020 has allowed advanced attackers to gain unauthorized access to files that may include customer data and intellectual property.
Darktrace does not use SolarWinds, and its operations remain unaffected by this breach. However, SolarWinds is an IT discovery tool that is used by a significant number of Darktrace customers. In what follows, we explore a set of Darktrace detections that highlight and alert security teams to the types of behaviors related to this breach.
This is not an example of a SolarWinds compromise, but examples of anomalous behaviors we can expect to see from this type of breach. These examples stress the value of self-learning Cyber AI capable of understanding the evolving normal ‘patterns of life’ within an enterprise – as opposed to a signature-based approach that looks at historical data to predict today’s threat.
As Darktrace detects device activity patterns rather than known malicious signatures, detecting use of these techniques will fall into the scope of Darktrace’s capabilities without further need for configuration. The technology automatically clusters devices into ‘peer groups’, allowing it to detect cases of an individual device behaving unusually. Using a self-learning approach is the best possible mechanism to catch an attacker who gains access into your systems using a degree of stealth so as to not trigger signature-based detection.
A number of these models may fire in combination with other models in order to make a strong detection over a time-series – and this is exactly where Darktrace’s autonomous incident triage capability, Cyber AI Analyst, plays a crucial role in investigating the alerts on behalf of security teams. Cyber AI Analyst saves critical time for security teams, and its results should be treated with a high priority during this period of vigilance.
How SolarWinds was detected with AI
We want to focus on the most sophisticated details of the hands-on intrusion that in many cases followed the initial automated attack. This post-exploitation part of the attack is much more varied and stealthy. These stages are also near-impossible to predict, as they are driven by the attacker’s intentions and goals for each individual victim at this stage – making the use of signatures, threat intelligence or static use cases virtually useless.
While the automated, initial malware execution is a critical initial step to understand, the behavior was pre-configured for the malware and included the download of further payloads and the connection to domain-generation-algorithm (DGA) based subdomains of avsvmcloud[.]com. These automated first stages of the attack have been sufficiently researched in depth by the community. This post is not aiming to add anything to these findings, but instead takes a look at the potential post-infection activities.
Malware / C2 domains
The threat-actor set the hostnames on their later-stage command and control (C2) infrastructure to match a legitimate hostname found within the victim’s environment. This allowed the adversary to blend into the environment, avoid suspicion, and evade detection. They further used C2 servers in geopolitical proximity to their victims, further circumventing static geo-based trusts lists. Darktrace is unaffected by this type of tradecraft as it does not have implicit, pre-defined trust of any geo-locations.
This would be very likely to trigger the following Darktrace Cyber AI models. The models were not specifically designed to detect SolarWinds modifications but have been in place for years – they are designed to detect the subtle but significant attacker activities occurring within an organization’s network.
Compromise / Agent Beacon to New Endpoint
Compromise / SSL Beaconing to New Endpoint
Compromise / HTTP Beaconing to New Endpoint*
*The implant uses SSL, but may be identified as HTTP if using a proxy.
Lateral movement using different credentials
Once the attacker gained access to the network with compromised credentials, they moved laterally using multiple different credentials. The credentials used for lateral movement were always different from those used for remote access.
This very likely would trigger the following Cyber AI models:
User / Multiple Uncommon New Credentials on Device
Figure 1: Example breach event log showing anomalous (new) logins from a single device, with multiple user credentials
User / New Admin Credentials on Client
Figure 2: Example breach event log showing anomalous admin login
Temporary file replacement and temporary task modification
The attacker used a temporary file replacement technique to remotely execute utilities: they replaced a legitimate utility with theirs, executed their payload, and then restored the legitimate original file. They similarly manipulated scheduled tasks by updating an existing legitimate task to execute their tools and then returned the scheduled task to its original configuration. They routinely removed their tools – including the removal of backdoors once legitimate remote access was achieved.
This would be very likely to trigger the following Cyber AI models:
Anomalous Connection / New or Uncommon Service Control
Figure 3: Example breach showing uncommon service control
Anomalous Connection / High Volume of New or Uncommon Service Control
Figure 4: Example breach showing 10 uncommon service controls
Device / AT Service Scheduled Task
Figure 5: Breach event log shows new AT service scheduled task activity
Device / Multiple RPC Requests for Unknown Services
Figure 6: Breach shows multiple binds to unknown RPC services
Device / Anomalous SMB Followed By Multiple Model Breaches
Figure 10: Breach shows significant deviation in SMB activity from device
SolarWinds breach remembered
By understanding where credentials are used and which devices talk to each other, Cyber AI has an unprecedented and dynamic understanding of business systems. This empowers it to alert security teams to enterprise changes that could indicate cyber risk in real time.
These alerts demonstrate how AI learns ‘normal’ for the unique digital environment surrounding it, and then alerts operators to deviations, including those that are directly relevant to the SUNBURST compromise. It further provides insights into how the attacker exploited those networks that did not have the appropriate visibility and detection capabilities.
On top of these alerts, Cyber AI Analyst will also be automatically correlating these detections over time to identify patterns, generating comprehensive and intuitive incident summaries and significantly reducing triage time. Reviewing Cyber AI Analyst alerts should be given high priority over the next several weeks.
Darktrace cyber analysts are world-class experts in threat intelligence, threat hunting and incident response, and provide 24/7 SOC support to thousands of Darktrace customers around the globe. Inside the SOC is exclusively authored by these experts, providing analysis of cyber incidents and threat trends, based on real-world experience in the field.
Atomic Stealer: Darktrace’s Investigation of a Growing macOS Threat
The Rise of Infostealers Targeting Apple Users
In a threat landscape historically dominated by Windows-based threats, the growing prevalence of macOS information stealers targeting Apple users is becoming an increasing concern for organizations. Infostealers are a type of malware designed to steal sensitive data from target devices, often enabling attackers to extract credentials and financial data for resale or further exploitation. Recent research identified infostealers as the largest category of new macOS malware, with an alarming 101% increase in the last two quarters of 2024 [1].
What is Atomic Stealer?
Among the most notorious is Atomic macOS Stealer (or AMOS), first observed in 2023. Known for its sophisticated build, Atomic Stealer can exfiltrate a wide range of sensitive information including keychain passwords, cookies, browser data and cryptocurrency wallets.
Originally marketed on Telegram as a Malware-as-a-Service (MaaS), Atomic Stealer has become a popular malware due to its ability to target macOS. Like other MaaS offerings, it includes services like a web panel for managing victims, with reports indicating a monthly subscription cost between $1,000 and $3,000 [2]. Although Atomic Stealer’s original intent was as a standalone MaaS product, its unique capability to target macOS has led to new variants emerging at an unprecedented rate
Even more concerning, the most recent variant has now added a backdoor for persistent access [3]. This backdoor presents a significant threat, as Atomic Stealer campaigns are believed to have reached an around 120 countries. The addition of a backdoor elevates Atomic Stealer to the rare category of backdoor deployments potentially at a global scale, something only previously attributed to nation-state threat actors [4].
This level of sophistication is also evident in the wide range of distribution methods observed since its first appearance; including fake application installers, malvertising and terminal command execution via the ClickFix technique. The ClickFix technique is particularly noteworthy: once the malware is downloaded onto the device, users are presented with what appears to be a legitimate macOS installation prompt. In reality, however, the user unknowingly initiates the execution of the Atomic Stealer malware.
This blog will focus on activity observed across multiple Darktrace customer environments where Atomic Stealer was detected, along with several indicators of compromise (IoCs). These included devices that successfully connected to endpoints associated with Atomic Stealer, those that attempted but failed to establish connections, and instances suggesting potential data exfiltration activity.
Darktrace’s Coverage of Atomic Stealer
As this evolving threat began to spread across the internet in June 2025, Darktrace observed a surge in Atomic Stealer activity, impacting numerous customers in 24 different countries worldwide. Initially, most of the cases detected in 2025 affected Darktrace customers within the Europe, Middle East, and Africa (EMEA) region. However, later in the year, Darktrace began to observe a more even distribution of cases across EMEA, the Americas (AMS), and Asia Pacific (APAC). While multiple sectors were impacted by Atomic Stealer, Darktrace customers in the education sector were the most affected, particularly during September and October, coinciding with the return to school and universities after summer closures. This spike likely reflects increased device usage as students returned and reconnected potentially compromised devices to school and campus environments.
Starting from June, Darktrace detected multiple events of suspicious HTTP activity to external connections to IPs in the range 45.94.47.0/24. Investigation by Darktrace’s Threat Research team revealed several distinct patterns ; HTTP POST requests to the URI “/contact”, identical cURL User Agents and HTTP requests to “/api/tasks/[base64 string]” URIs.
Within one observed customer’s environment in July, Darktrace detected two devices making repeated initiated HTTP connections over port 80 to IPs within the same range. The first, Device A, was observed making GET requests to the IP 45.94.47[.]158 (AS60781 LeaseWeb Netherlands B.V.), targeting the URI “/api/tasks/[base64string]” using the “curl/8.7.2” user agent. This pattern suggested beaconing activity and triggered the ‘Beaconing Activity to External Rare' model alert in Darktrace / NETWORK, with Device A’s Model Event Log showing repeated connections. The IP associated with this endpoint has since been flagged by multiple open-source intelligence (OSINT) vendors as being associated with Atomic Stealer [5].
Figure 1: Darktrace’s detection of Device A showing repeated connections to the suspicious IP address over port 80, indicative of beaconing behavior.
Darktrace’s Cyber AI Analyst subsequently launched an investigation into the activity, uncovering that the GET requests resulted in a ‘503 Service Unavailable’ response, likely indicating that the server was temporarily unable to process the requests.
Figure 2: Cyber AI Analyst Incident showing the 503 Status Code, indicating that the server was temporarily unavailable.
This unusual activity prompted Darktrace’s Autonomous Response capability to recommend several blocking actions for the device in an attempt to stop the malicious activity. However, as the customer’s Autonomous Response configuration was set to Human Confirmation Mode, Darktrace was unable to automatically apply these actions. Had Autonomous Response been fully enabled, these connections would have been blocked, likely rendering the malware ineffective at reaching its malicious command-and-control (C2) infrastructure.
Figure 3: Autonomous Response’s suggested actions to block suspicious connectivity on Device A in the first customer environment.
In another customer environment in August, Darktrace detected similar IoCs, noting a device establishing a connection to the external endpoint 45.94.47[.]149 (ASN: AS57043 Hostkey B.V.). Shortly after the initial connections, the device was observed making repeated requests to the same destination IP, targeting the URI /api/tasks/[base64string] with the user agent curl/8.7.1, again suggesting beaconing activity. Further analysis of this endpoint after the fact revealed links to Atomic Stealer in OSINT reporting [6].
Figure 4: Cyber AI Analyst investigation finding a suspicious URI and user agent for the offending device within the second customer environment.
As with the customer in the first case, had Darktrace’s Autonomous Response been properly configured on the customer’s network, it would have been able to block connectivity with 45.94.47[.]149. Instead, Darktrace suggested recommended actions that the customer’s security team could manually apply to help contain the attack.
Figure 5: Autonomous Response’s suggested actions to block suspicious connectivity to IP 45.94.47[.]149 for the device within the second customer environment.
In the most recent case observed by Darktrace in October, multiple instances of Atomic Stealer activity were seen across one customer’s environment, with two devices communicating with Atomic Stealer C2 infrastructure. During this incident, one device was observed making an HTTP GET request to the IP 45.94.47[.]149 (ASN: AS60781 LeaseWeb Netherlands B.V.). These connections targeted the URI /api/tasks/[base64string, using the user agent curl/8.7.1.
Shortly afterward, the device began making repeated connections over port 80 to the same external IP, 45.94.47[.]149. This activity continued for several days until Darktrace detected the device making an HTTP POST request to a new IP, 45.94.47[.]211 (ASN: AS57043 Hostkey B.V.), this time targeting the URI /contact, again using the curl/8.7.1 user agent. Similar to the other IPs observed in beaconing activity, OSINT reporting later linked this one to information stealer C2 infrastructure [7].
Figure 6: Darktrace’s detection of suspicious beaconing connectivity with the suspicious IP 45.94.47.211.
Further investigation into this customer’s network revealed that similar activity had been occurring as far back as August, when Darktrace detected data exfiltration on a second device. Cyber AI Analyst identified this device making a single HTTP POST connection to the external IP 45.94.47[.]144, another IP with malicious links [8], using the user agent curl/8.7.1 and targeting the URI /contact.
Figure 7: Cyber AI Analyst investigation finding a successful POST request to 45.94.47[.]144 for the device within the third customer environment.
A deeper investigation into the technical details within the POST request revealed the presence of a file named “out.zip”, suggesting potential data exfiltration.
Figure 8: Advanced Search log in Darktrace / NETWORK showing “out.zip”, indicating potential data exfiltration for a device within the third customer environment.
Similarly, in another environment, Darktrace was able to collect a packet capture (PCAP) of suspected Atomic Stealer activity, which revealed potential indicators of data exfiltration. This included the presence of the “out.zip” file being exfiltrated via an HTTP POST request, along with data that appeared to contain details of an Electrum cryptocurrency wallet and possible passwords.
Read more about Darktrace’s full deep dive into a similar case where this tactic was leveraged by malware as part of an elaborate cryptocurrency scam.
Figure 9: PCAP of an HTTP POST request showing the file “out.zip” and details of Electrum Cryptocurrency wallet.
Although recent research attributes the “out.zip” file to a new variant named SHAMOS [9], it has also been linked more broadly to Atomic Stealer [10]. Indeed, this is not the first instance where Darktrace has seen the “out.zip” file in cases involving Atomic Stealer either. In a previous blog detailing a social engineering campaign that targeted cryptocurrency users with the Realst Stealer, the macOS version of Realst contained a binary that was found to be Atomic Stealer, and similar IoCs were identified, including artifacts of data exfiltration such as the “out.zip” file.
Conclusion
The rapid rise of Atomic Stealer and its ability to target macOS marks a significant shift in the threat landscape and should serve as a clear warning to Apple users who were traditionally perceived as more secure in a malware ecosystem historically dominated by Windows-based threats.
Atomic Stealer’s growing popularity is now challenging that perception, expanding its reach and accessibility to a broader range of victims. Even more concerning is the emergence of a variant embedded with a backdoor, which is likely to increase its appeal among a diverse range of threat actors. Darktrace’s ability to adapt and detect new tactics and IoCs in real time delivers the proactive defense organizations need to protect themselves against emerging threats before they can gain momentum.
Credit to Isabel Evans (Cyber Analyst), Dylan Hinz (Associate Principal Cyber Analyst) Edited by Ryan Traill (Analyst Content Lead)
How Darktrace is ending email security silos with new capabilities in cross-domain detection, DLP, and native Microsoft integrations
A new era of reputation-aware, unified email security
Darktrace / EMAIL is redefining email defense with new innovations that close email security silos and empower SOC teams to stop multi-stage attacks – without disrupting business operations.
By extending visibility across interconnected domains, Darktrace catches the 17% of threats that leading SEGs miss, including multi-stage attacks like email bombing and cloud platform abuse. Its label-free behavioral DLP protects sensitive data without reliance on manual rules or classification, while DMARC strengthens brand trust and authenticity. With native integrations for Microsoft Defender and Security Copilot, SOC teams can now investigate and respond faster, reducing risk and maintaining operational continuity across the enterprise.
Summary of what’s new:
Cross-domain AI-native detection unifying email, identity, and SaaS
Label-free behavioral DLP for effortless data protection
Microsoft Defender and Security Copilot integrations for streamlined investigation and response
Why email security must evolve
Today’s attacks don’t stop at the inbox. They move across domains – email to identity, SaaS, and network – exploiting the blind spots between disconnected tools. Yet most email security solutions still operate in isolation, unable to see or respond beyond the message itself.
Tool sprawl compounds the issue. The average enterprise manages around 75 security products, and 69% report operational strain as a result. This complexity is counterproductive – and with legacy SEGs failing to adapt to detect threats that exploit human behavior, analysts are left juggling an unwieldy patchwork of fragmented defenses.
The bottom line? Siloed email defenses can’t keep pace with today’s AI-driven, cross domain attacks.
Beyond detection: AI built for modern threats
Darktrace / EMAIL is uniquely designed to catch the threats SEGs miss, powered by Self-Learning AI. It learns the communication patterns of every user – correlating behavioral signals from email, identity, and SaaS – to identify the subtle, context-driven deviations that define advanced social engineering and supply chain attacks.
Unlike tools that rely on static rules or historical attack data, Darktrace’s AI assumes a zero trust posture, treating every interaction as a potential risk. It detects novel threats in real time, including those that exploit trusted relationships or mimic legitimate business processes. And because Darktrace’s technology is natively unified, it delivers precise, coordinated responses that neutralize threats in real time.
Powerful innovations to Darktrace / EMAIL
Improved, multi-domain threat detection and response
With this update, Darktrace reveals multi-domain detection linking behavioral signals across email, identity, and SaaS to uncover advanced attacks. Darktrace leverages its existing agentic platform to understand behavioral deviations in any communication channel and take precise actions regardless of the domain.
This innovation enables customers to:
Correlate behavioral signals across domains to expose cross-channel threats and enable coordinated response
Link email and identity intelligence to neutralize multi-stage attacks, including advanced email bombing campaigns
Detection accuracy is further strengthened through layering with traditional threat intelligence:
Integrated antivirus verdicts improve detection efficacy by adding traditional file scanning
Structured threat intelligence (STIX/TAXII) enriches alerts with global context for faster triage and prioritization
Expanded ecosystem visibility also includes:
Salesforce integration, enabling automatic action on potentially malicious tickets auto-created from emails – accelerating threat response and reducing manual burden
Advancements in label-free DLP
Darktrace is delivering the industry’s first label-free data loss prevention (DLP) solution powered by a proprietary domain specific language model (DSLM).
This update expands DLP to protect against both secrets and personally identifiable information (PII), safeguarding sensitive data without relying on status rules or manual classification. The DSLM is tuned for email/DLP semantics so it understands entities, PII patterns, and message context quickly enough to enforce at send time.
Key enhancements include:
Behaviorally enhanced PII detection that automatically defines over 35+ new categories, including personal, financial, and health data
Added detail to DLP alerts in the UI, showing exactly how and when DLP policies were applied
Enhanced Cyber AI Analyst narratives to explain detection logic, making it easier to investigate and escalate incidents
And for further confidence in outbound mail, discover new updates to DMARC, with support for BIMI logo verification, automatic detection of both MTA-STS and TLS records, and data exports for deeper analysis and reporting. Accessible for all organizations, available now on the Azure marketplace.
Streamlined SOC workflows, with Microsoft-native integrations
This update introduces new integrations that simplify SOC operations, unify visibility, and accelerate response. By embedding directly into the Microsoft ecosystem – with Defender and Security Copilot – analysts gain instant access to correlated insights without switching consoles.
New innovations include:
Unified quarantine management with Microsoft Defender, centralizing containment within the native Microsoft interface and eliminating console hopping
Ability to surface threat insights directly in Copilot via the Darktrace Email Analysis Agent, eliminating data hunting and simplifying investigations
Automatic ticket creation in JIRA when users report suspicious messages
Sandbox analysis integration, enabling payload inspection in isolated environments directly from the Darktrace UI
Committed to innovation
These updates are part of the broader Darktrace release, which also included:
As attackers exploit gaps between tools, the Darktrace ActiveAI Security Platform delivers unified detection, automated investigation, and autonomous response across cloud, endpoint, email, network, and OT. With full-stack visibility and AI-native workflows, Darktrace empowers security teams to detect, understand, and stop novel threats before they escalate.
Join our Live Launch Event
When? December 9, 2025
What will be covered? Join our live broadcast to experience how Darktrace is eliminating blind spots for detection and response across your complete enterprise with new innovations in Agentic AI across our ActiveAI Security platform. Industry leaders from IDC will join Darktrace customers to discuss challenges in cross-domain security, with a live walkthrough reshaping the future of Network Detection & Response, Endpoint Detection & Response, Email Security, and SecOps in novel threat detection and autonomous investigations.
.jpg)
![Autonomous Response’s suggested actions to block suspicious connectivity to IP 45.94.47[.]149 for the device within the second customer environment.](https://cdn.prod.website-files.com/626ff4d25aca2edf4325ff97/6931e3cb1ddc36af8205720b_Screenshot%202025-12-04%20at%2011.40.54%E2%80%AFAM.png)
![Cyber AI Analyst investigation finding a successful POST request to 45.94.47[.]144 for the device within the third customer environment.](https://cdn.prod.website-files.com/626ff4d25aca2edf4325ff97/6931e41d22eb7e9a3e8217d5_Screenshot%202025-12-04%20at%2011.42.15%E2%80%AFAM.png)
