Blog
/
Network
/
August 5, 2025

2025 Cyber Threat Landscape: Darktrace’s Mid-Year Review

Explore key cyber threat trends observed across Darktrace’s customer base in the first half of 2025. As threat actors increasingly adopt AI and diversify their techniques and tooling, anomaly-based detection continues to prove vital in defending against evolving attacks.
Inside the SOC
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
Emma Foulger
Global Threat Research Operations Lead
cyberseucity 2025 half year threat report Default blog image
05
Aug 2025

Note: Following initial publication, an error was identified in the previously reported number of QR codes observed in phishing emails in February 2025. This figure was originally stated as over 1 million, when in fact it was over 100,000. The error has since been corrected.

‍

‍

2025: Threat landscape in review

The following is a retrospective of the first six months of 2025, highlighting key findings across the threat landscape impacting Darktrace customers.

Darktrace observed a wide range of tactics during this period, used by various types of threat actors including advanced persistent threats (APTs), Malware-as-a-Service (MaaS) and Ransomware-as-a-Service (RaaS) groups.

Methodology

Darktrace’s Analyst team conduct investigations and research into threats facing organizations and security teams across our customer base.  This includes direct investigations with our 24/7 Security Operations Centre (SOC), via services such as Managed Detection and Response (MDR) and Managed Threat Detection, as well as broader cross-fleet research through our Threat Research function.

At the core of our research is Darktrace’s anomaly-based detection, which the Analyst team contextualizes and analyzes to provide additional support to customers and deepen our understanding of the threats they face.

Threat actors are incorporating AI into offensive operations

Threat actors are continuously evolving their tactics, techniques, and procedures (TTPs), posing an ongoing challenge to effective defense hardening. Increasingly, many threat actors are adopting AI, particularly large language models (LLMs), into their operations to enhance the scale, sophistication, and efficacy of their attacks.

The evolving functionality of malware, such as the recently reported LameHug malware by CERT-UA, which uses an open-source LLM, exemplifies this observation [1].

Threat landscape trends in 2025

Threat actors applying AI to Email attacks

LLMs present a clear opportunity for attackers to take advantage of AI and create effective phishing emails at speed. While Darktrace cannot definitively confirm the use of AI to create the phishing emails observed across the customer base, the high volume of phishing emails and notable shifts in tactic could potentially be explained by threat actors adopting new tooling such as LLMs.

  • The total number of malicious emails detected by Darktrace from January to May 2025 was over 12.6 million
  • VIP users continue to face significant threat, with over 25% of all phishing emails targeting these users in the first five months of 2025 ‍
  • QR code-based phishing emails have remained a consistent tactic, with a similar proportion observed in January-May 2024 and 2025. The highest numbers were observed in February 2025, with over 100,000 detected in that month alone.
  • Shifts towards increased sophistication within phishing emails are emerging, with a year-on-year increase in the proportion of phishing emails containing either a high text volume or multistage payloads. In the first five months of 2025, 32% of phishing emails contained a high volume of text.

The increase in proportion of phishing emails with a high volume of text in particular could point towards threat actors leveraging LLMs to create phishing emails with large, but believable, text in an easy and efficient way.

The above email statistics are derived from analysis of monitored Darktrace / EMAIL model data for all customer deployments hosted in the cloud between January 1 and May 31, 2025.

Campaign Spotlight: Simple, Quick - ClickFix

An interesting technique Darktrace observed multiple times throughout March and April was ClickFix social engineering, which exploits the intersection between humans and technology to trick users into executing malicious code on behalf of the attacker.

  • While this technique has been around since 2024, Darktrace observed campaign activity in the first half of 2025 suggesting a resurgence.  
  • A range of threat actors – from APTs to MaaS and RaaS have adopted this technique to deliver secondary payloads, like information stealing malware.
  • Attackers use fraudulent or compromised legitimate websites to inject malicious plugins that masquerade as fake CAPTCHAs.
  • Targeted users believe they are completing human verification or resolving a website issue, unaware that they are being guided through a series of simple steps to execute PowerShell code on their system.
  • Darktrace observed campaign activity during the first half of 2025 across a range of sectors, including Government, Healthcare, Insurance, Retail and, Non-profit.

Not just AI: Automation is enabling Ransomware and SaaS exploitation

The rise of phishing kits like FlowerStorm and Mamba2FA, which enable phishing and abuse users’ trust by mimicking legitimate services to bypass multi-factor authentication (MFA), highlight how the barriers to entry for sophisticated attacks continue to fall, enabling new threat actors. Combined with Software-as-a-Service (SaaS) account compromise, these techniques make up a substantial portion of cybercriminal activity observed by Darktrace so far this year.

Credentials remain the weak link

A key theme across multiple cases of ransomware was threat actors abusing compromised credentials to gain initial entry into networks via:

  • Unauthorized access to internet-facing technology such as RDP servers and virtual private networks (VPNs).
  • Unauthorized access to SaaS accounts.

SaaS targeted ransomware is on the rise

The encryption of files within SaaS environments observed by Darktrace demonstrates a continued trend of ransomware actors targeting these platforms over traditional networks, potentially driven by a higher return on investment.

SaaS accounts are often less protected than traditional systems because of Single Sign-On (SSO).  Additionally, platforms like Salesforce often host sensitive data, including emails, financial records, customer information, and network configuration details. This stresses the need for robust identity management practices and continuous monitoring.

RaaS is adding complexity and speed to cyber attacks

RaaS has dominated the attack landscape, with groups like Qilin, RansomHub, and Lynx all appearing multiple times in cases across Darktrace’s customer base over the past six months. Detecting ransomware attacks before the encryption stage remains a significant challenge, particularly in RaaS operations where different affiliates often use varying techniques for initial entry and earlier stages of the attack. Darktrace’s recent analysis of Scattered Spider underscores the challenge of hardening defenses against such varying techniques.

CVE exploitation continues despite available patches

Darktrace has also observed ransomware gangs exploiting known Common Vulnerabilities and Exposures (CVEs), including the Medusa ransomware group’s use of the SimpleHelp vulnerabilities: CVE-2024-57727 and CVE-2024-57728 in March, despite patches being made available in January [2].

Misused tools + delayed patches = growing cyber risk

The exploitation of common remote management tools like SimpleHelp highlights the serious challenges defenders face when patch management cycles are suboptimal. As threat actors continue to abuse legitimate services for malicious purposes, the challenges facing defenders will only grow more complex.

Edge exploitation

It comes as no surprise that exploitation of internet-facing devices continued to feature prominently in Darktrace’s Threat Research investigations during the first half of 2025.

Observed CVE exploitation included:

Many of Darktrace’s observations of CVE exploitation so far in 2025 align with wider industry reporting, which suggests that Chinese-nexus threat actors were deemed to likely have exploited these technologies prior to public disclosure. In the case of CVE-2025-0994 - a vulnerability affecting Trimble Cityworks, an asset management system designed for use by local governments, utilities, airports, and public work agencies [3] - Darktrace observed signs of exploitation as early as January 19, well before vulnerability’s public disclosure on February 6 [4]. Darktrace’s early identification of the exploitation stemmed from the detection of a suspicious file download from 192.210.239[.]172:3219/z44.exe - later linked to Chinese-speaking threat actors in a campaign targeting the US government [5].

This case demonstrates the risks posed by the exploitation of internet-facing devices, not only those hosting more common technologies, but also software associated specifically tied to Critical National Infrastructure (CNI); a lucrative target for threat actors. This also highlights Darktrace’s ability to detect exploitation of internet-facing systems, even without a publicly disclosed CVE. Further examples of how Darktrace’s anomaly detection can uncover malicious activity ahead of public vulnerability disclosures can be found here.

New threats and returning adversaries

In the first half of 2025, Darktrace observed a wide range of threats, from sophisticated techniques employed by APT groups to large-scale campaigns involving phishing and information stealers.

BlindEagle (APT-C-36)

Among the observed APT activity, BlindEagle (APT-C-36) was seen targeting customers in Latin America (LATM), first identified in February, with additional cases seen as recently as June.

Darktrace also observed a customer targeted in a China-linked campaign involving the LapDogs ORB network, with activity spanning from December 2024 and June 2025. These likely nation-state attacks illustrate the continued adoption of cyber and AI capabilities into the national security goals of certain countries.

Sophisticated malware functionality

Further sophistication has been observed within specific malware functionality - such as the malicious backdoor Auto-Color, which has now been found to employ suppression tactics to cover its tracks if it is unable to complete its kill chain - highlighting the potential for advanced techniques across every layer of an attack.‍

Familiar foes

Alongside new and emerging threats, previously observed and less sophisticated tools, such as worms, Remote Access Trojans (RATs), and information stealers, continue to impact Darktrace customers.

The Raspberry Robin worm... First seen in 2021, has been repeatedly identified within Darktrace’s customer base since 2022. Most recently, Darktrace’s Threat Research team identified cases in April and May this year. Recent open-source intelligence (OSINT) reporting suggests that Raspberry Robin continues to evolve its role as an Initial Access Broker (IAB), paving the way for various attacks and remaining a concern [6].

RATs also remain a threat, with examples like AsyncRAT and Gh0st RAT impacting Darktrace customers.

In April multiple cases of MaaS were observed in Darktrace’s customer base, with information stealers Amadey and Stealc, as well as GhostSocks being distributed as a follow up payload after an initial Amadey infection.

Conclusion

As cyber threats evolve, attackers are increasingly harnessing AI to craft highly convincing email attacks, automating phishing campaigns at unprecedented scale and speed. This, coupled with rapid exploitation of vulnerabilities and the growing sophistication of ransomware gangs operating as organized crime syndicates, makes today’s threat landscape more dynamic and dangerous than ever. Cyber defenders collaborate to combat these threats – the coordinated takedown of Lumma Stealer in May was a notable win for both industry and law-enforcement [7], however OSINT suggests that this threat persists [8], and new threats will continue to arise.

Traditional security tools that rely on static rules or signature-based detection often struggle to keep pace with these fast-moving, adaptive threats. In this environment, anomaly-based detection tools are no longer optional—they are essential. By identifying deviations in normal user and system behavior, tools like Darktrace provide a proactive layer of defense capable of detecting novel and emerging threats, even those that bypass conventional security measures. Investing in anomaly-based detection is critical to staying ahead of attackers who now operate with automation, intelligence, and global coordination.

Credit to Emma Foulger (Global Threat Research Operations Lead), Nathaniel Jones (VP, Security & AI Strategy, Field CISO),  Eugene Chua (Principal Cyber Analyst & Analyst Team Lead), Nahisha Nobregas (Senior Cyber Analyst), Nicole Wong (Principal Cyber Analyst), Justin Torres (Senior Cyber Analyst), Matthew John (Director of Operations, SOC), Sam Lister (Specialist Security Researcher), Ryan Traill (Analyst Content Lead) and the Darktrace Incident Management team.

The information contained in this blog post is provided for general informational purposes only and represents the views and analysis of Darktrace as of the date of publication. While efforts have been made to ensure the accuracy and timeliness of the information, the cybersecurity landscape is dynamic, and new threats or vulnerabilities may have emerged since this report was compiled.

This content is provided “as is” and without warranties of any kind, either express or implied. Darktrace makes no representations or warranties regarding the completeness, accuracy, reliability, or suitability of the information, and expressly disclaims all warranties.

Nothing in this blog post should be interpreted as legal, technical, or professional advice. Users of this information assume full responsibility for any actions taken based on its content, and Darktrace shall not be liable for any loss or damage resulting from reliance on this material. Reference to any specific products, companies, or services does not constitute or imply endorsement, recommendation, or affiliation.

Appendices

Indicators of Compromise (IoCs)

IoC - Type - Description + Probability

LapDogs ORB network, December 2024-June 2025

www.northumbra[.]com – Hostname – Command and Control (C2) server

103.131.189[.]2 – IP Address - C2 server, observed December 2024 & June 2025

103.106.230[.]31 – IP Address - C2 server, observed December 2024

154.223.20[.]56 – IP Address – Possible C2 server, observed December 2024

38.60.214[.]23 – IP Address – Possible C2 server, observed January & February 2025

154.223.20[.]58:1346/systemd-log – URL – Possible ShortLeash payload, observed December 2024

CN=ROOT,OU=Police department,O=LAPD,L=LA,ST=California,C=US - TLS certificate details for C2 server

CVE-2025-0994, Trimble Cityworks exploitation, January 2025

192.210.239[.]172:3219/z44.exe – URL - Likely malicious file download

AsyncRAT, February-March 2025

windows-cam.casacam[.]net – Hostname – Likely C2 server

88.209.248[.]141 – IP Address – Likely C2 server

207.231.105[.]51 – IP Address – Likely C2 server

163.172.125[.]253 – IP Address – Likely C2 server

microsoft-download.ddnsfree[.]com – Hostname – Likely C2 server

95.217.34[.]113 – IP Address – Likely C2 server

vpnl[.]net – Hostname – Likely C2 server

157.20.182[.]16 – IP Address - Likely C2 server

185.81.157[.]19 – IP Address – Likely C2 server

dynamic.serveftp[.]net – IP Address – Likely C2 server

158.220.96.15 – IP Address – Likely C2 server

CVE-2024-57727 & CVE-2024-57728, SimpleHelp RMM exploitation, March 2025

213.183.63[.]41 – IP Address - C2 server

213.183.63[.]41/access/JWrapper-Windows64JRE-version.txt?time=3512082867 – URL - C2 server

213.183.63[.]41/access/JWrapper-Windows64JRE-00000000002-archive.p2.l2 – URL - C2 server

pruebas.pintacuario[.]mx – Hostname – Possible C2 server

144.217.181[.]205 – IP Address – Likely C2 server

erp.ranasons[.]com – Hostname – Possible destination for exfiltration

143.110.243[.]154 – IP Address – Likely destination for exfiltration

Blind Eagle, April-June 2025

sostenermio2024.duckdns[.]org/31agosto.vbs – URL – Possible malicious file download

Stealc, April 2025

88.214.48[.]93/ea2cb15d61cc476f[.]php – URL – C2 server

Amadey & GhostSocks, April 2025

195.82.147[.]98 – IP Address - Amadey C2 server

195.82.147[.]98/0Bdh3sQpbD/index.php – IP Address – Likely Amadey C2 activity

194.28.226.181 – IP Address – Likely GhostSocks C2 server

RaspberryRobin, May 2025

4j[.]pm – Hostname – C2 server

4xq[.]nl – Hostname – C2 server

8t[.]wf – Hostname – C2 server

Gh0stRAT, May 2025

lu.dssiss[.]icu  - Hostname – Likely C2 server

192.238.133[.]162:7744/1-111.exe – URL – Possible addition payload

8e9dec3b028f2406a8c546a9e9ea3d50609c36bb - SHA1 - Possible additional payload

f891c920f81bab4efbaaa1f7a850d484 - MD5 – Possible additional payload

192.238.133[.]162:7744/c3p.exe – URL - Possible additional payload

03287a15bfd67ff8c3340c0bae425ecaa37a929f - SHA1 - Possible additional payload

02aa02aee2a6bd93a4a8f4941a0e6310 - MD5 - Possible additional payload

192.238.133[.]162:7744/1-1111.exe – URL - Possible additional payload

1473292e1405882b394de5a5857f0b6fa3858fd1 - SHA1 - Possible additional payload

69549862b2d357e1de5bab899ec0c817 - MD5 - Possible additional payload

192.238.133[.]162:7744/1-25.exe – URL -  Possible additional payload

20189164c4cd5cac7eb76ba31d0bd8936761d7a7  - SHA1 - Possible additional payload

f42aa5e68b28a3f335f5ea8b6c60cb57 – MD5 - Possible additional payload

192.238.133[.]162:7744/Project1_se.exe – URL - Possible additional payload

fea1e30dfafbe9fa9abbbdefbcbe245b6b0628ad - SHA1 - Possible additional payload

5ea622c630ef2fd677868cbe8523a3d5 - MD5 - Possible additional payload

192.238.133[.]162:7744/Project1_se.exe - URL - Possible additional payload

aa5a5d2bd610ccf23e58bcb17d6856d7566d71b9  - SHA1 - Possible additional payload

9d33029eaeac1c2d05cf47eebb93a1d0 - MD5 - Possible additional payload

References and further reading

1.        https://cip.gov.ua/en/news/art28-atakuye-sektor-bezpeki-ta-oboroni-za-dopomogoyu-programnogo-zasobu-sho-vikoristovuye-shtuchnii-intelekt?utm_medium=email&_hsmi=113619842&utm_content=113619842&utm_source=hs_email

2.        https://www.s-rminform.com/latest-thinking/cyber-threat-advisory-medusa-and-the-simplehelp-vulnerability

3.        https://assetlifecycle.trimble.com/en/products/software/cityworks

4.     https://nvd.nist.gov/vuln/detail/CVE-2025-0994

5.     https://blog.talosintelligence.com/uat-6382-exploits-cityworks-vulnerability/

6.        https://www.silentpush.com/blog/raspberry-robin/

7.        https://blogs.microsoft.com/on-the-issues/2025/05/21/microsoft-leads-global-action-against-favored-cybercrime-tool/

8.     https://www.trendmicro.com/en_sg/research/25/g/lumma-stealer-returns.html

Related Darktrace investigations

-              ClickFix

-              FlowerStorm

-              Mamba 2FA

-              Qilin Ransomware

-              RansomHub Ransomware

-              RansomHub Revisited

-              Lynx Ransomware

-              Scattered Spider

-              Medusa Ransomware

-              Legitimate Services Malicious Intentions

-              CVE-2025-0282 and CVE-2025-0283 – Ivanti CS, PS and ZTA

-              CVE-2025-31324 – SAP Netweaver

-              Pre-CVE Threat Detection

-              BlindEagle (APT-C-36)

-              Raspberry Robin Worm

-              AsyncRAT

-              Amadey

-              Lumma Stealer

Inside the SOC
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
Emma Foulger
Global Threat Research Operations Lead

More in this series

No items found.

Blog

/

AI

/

November 18, 2026

Microsoft Perception, Agentic Insiders and the Role of Behavioral Security

Default blog imageDefault blog image

This week at Ignite, Microsoft announced that Microsoft Perception is generally available. It is a significant milestone for the industry, and one we welcome as a Microsoft partner. Microsoft is putting an architectural bet at the center of its security strategy: AI has changed the speed of cyber, and defense must change with it.

At Darktrace, we’ve long held the same view, so it’s encouraging to see a partner of Microsoft’s scale articulate it so clearly and bring it to market.

Security gets a team of agents

Perception is Microsoft’s agentic security system. It orchestrates red, blue, and green team agents through playbooks in a closed loop: red agents simulate attacks, blue agents detect, triage and investigate, and green agents fix and remediate. Rather than adding another assistant to the SOC, it creates coordinated teams of agents that share context. Security teams set the guardrails and approve high-impact actions.

Microsoft frames this as a three-layer cyber stack: signals, sensors and actuators; a context layer; and agents, harness and models on top. Together with partner sources, the foundation produces more than 165 trillion signals per day.  

The announcement also makes a second point that matters just as much. Microsoft is treating AI agents themselves as a new insider-risk category. In its words, “Agents are the new insiders.” They work at machine speed, around the clock, with broad access to data, apps, tools and other agents, and like any insider they can make mistakes, be manipulated or be compromised. New capabilities such as Dynamic Agent Risk (now in preview) and dynamic Conditional Access controls are designed to assess and contain that risk as it changes.

Four truths about security in the agentic era

The announcement is built on four ideas, and together they describe the new reality security teams are defending:

  1. Static security can’t keep up. Microsoft validates what many security practitioners have observed; the window between a vulnerability’s discovery and its exploitation is collapsing from months to minutes. Manual triage can’t answer that pace. Autonomous detection and response is now a requirement, not simply a “nice to have”.
  2. Agents must be treated as insiders. The question for security teams is no longer only “who is this user?” but “what is this agent actually doing?” Identity tells you who an agent is. Policy tells you what it is allowed to do. Neither tells you whether its behavior is normal.
  3. Security is moving from signatures to behavior and context. Dynamic Agent Risk looks across identity, data, endpoint, cloud and network signals and asks whether an agent’s actions are drifting from the intent of its task. That is behavioral thinking, and it is philosophically close to the Adaptive AI approach Darktrace has pioneered.
  4. Machine-speed attacks demand machine-speed response. Humans set the objectives and guardrails. Autonomous systems do the work inside them, both detecting and acting. Microsoft’s design has agents investigate, recommend and, within approved boundaries, act, with security teams setting the limits. At this pace, the gap between knowing something is wrong and doing something about it has to close to seconds.

The opportunity: where control plane meets behavioral context

Perception is an extraordinary control plane for agentic security. The question for security leaders is how to make it even more effective in the environments they actually run.

Real enterprises are heterogeneous. They run AWS and Google Cloud alongside Azure, and OT, IoT and third-party identity alongside Entra. Increasingly they also run AI agents and MCP-connected systems that no single vendor’s telemetry fully covers. Microsoft’s architecture recognizes this, with more than 500 connectors into the wider ecosystem.

That is where Darktrace’s role is distinct. The challenge isn’t signal volume, where Microsoft’s scale is unmatched. It’s deeply understanding what is normal for a specific organization, across every system, identity and agent in it. Together, we deliver two complementary layers:

  • Microsoft brings global intelligence and orchestration. It tells you what is known, what it means and what to do about it.
  • Darktrace brings organization-specific behavioral context. It tells you what is unusual here, right now, across the organizations AI, people and infrastructure, even if it has never been seen before.

The difference shows up in the numbers. A large enterprise generates billions of signals. Darktrace turns them into thousands of incidents, narrows those to hundreds of AI-led investigations, and surfaces as few as 20 that merit a human’s attention. For an agentic SOC, that kind of evaluated, prioritized input makes agents faster, not busier.

As enterprises deploy more agents, the hard problem becomes behavior that is technically permitted but unusual, manipulated or inconsistent with the agent’s purpose. Consider an agent that normally reads Salesforce records during US business hours and suddenly starts enumerating SharePoint and opening unusual outbound connections. Each step may use valid access and approved tools, so identity and policy checks can all pass. Spotting it requires knowing the behavior is abnormal for this agent, in this environment, right now.

That is why one principle in Microsoft’s AI Transparency Initiative matters so much: organizations should be able to see what their AI accesses and does, and independently verify those actions.  

  • Identity and permissions answer “should this agent be allowed to do this?”
  • Behavioral defense answers “even though it is allowed, is what it’s doing normal?”

Deterministic controls around non-deterministic AI are the right foundation, and continuously learned baselines are how organizations can verify those controls are working. ‍

What to look for in an agentic SOC architecture

As organizations adopt platforms like Perception, we’d encourage them to look for three capabilities:

  1. Behavioral context in the control plane. Agents work best with high-confidence signals rather than thousands of raw alerts: here is the behavior that materially deviates from this organization’s learned normal.
  2. ‍Behavioral reasoning on demand. During an investigation, agents should be able to ask what changed around a given identity, device or agent in the last 24 hours, and get a clear answer. ‍
  3. Autonomous, coordinated response across the stack. Response should be as fast as the threat without being reckless: acting immediately when confidence is high, constraining or reversing actions when it isn’t, and keeping a human in the loop when it matters. Within policy and human-approval guardrails, it should target the abnormal behavior rather than the whole business process and span identity, endpoint and the wider environment.

These are the areas where we’re focused, and where we look forward to working with Microsoft and the broader ecosystem.

Response is where speed is won or lost

In the AI era, detection and investigation only matter if action follows quickly. Perception brings agentic orchestration to that problem, with security teams setting the guardrails and approving high-impact steps. Autonomous response from Darktrace takes a complementary approach: it responds on behavioral deviation, in a way that is proportionate and targeted. It can interrupt activity that is abnormal for that specific device, identity or agent while leaving normal business activity running.

The two approaches fit together naturally. Behavioral signals can tell an orchestration layer that something has changed and how confident the system is. Orchestrated playbooks can then coordinate wider containment across identity, endpoint and cloud, within the policies customers define.

The bigger picture

The industry’s architecture is converging on identity, policy, telemetry, behavior, autonomous reasoning and machine-speed response. Microsoft is building formidable pieces of that picture, especially identity, policy, telemetry and orchestration.

Behavior is the layer Darktrace is built to provide: an independent intelligence that continuously learns how humans, machines and AI agents normally act, detects the moment reality departs from intent, and can respond to that deviation in a targeted, proportionate way that keeps normal business running. When attacks move at machine speed, understanding what is abnormal and acting on it quickly have to go together.

Perception brings together what Microsoft knows and orchestrates the response across its stack. Darktrace helps organizations understand what is normal for them and act on what isn’t. Customers building their agentic security strategy will need both.

Congratulations to our partners at Microsoft on this milestone. We look forward to helping customers secure the agentic era together.

Learn more about Darktrace Behavioral Defense.

Continue reading
About the author
Kellie Regan
Director, Product Marketing - Cloud Security

Blog

/

AI

/

November 18, 2026

Agentic attacks in the wild: Inside three ARTEX investigations

Default blog imageDefault blog image

Key insights

·       Darktrace observed activity in several customer environments involving infrastructure and behaviors associated with reported ARTEX-enabled campaigns, including activity linked to China-nexus espionage and financially motivated attacks.

·       In the higher-confidence cases, known infrastructure appeared alongside beaconing, unusual TLS, reconnaissance, tunneling, script retrieval, SMB activity and external data transfer. One Azure case had no observed network command and control at all, only key listing and application identity activity.

·       ARTEX is now used by actors with different objectives. Identifying the framework can help locate relevant activity, while the behavior around the affected system provides the stronger evidence for understanding compromise and attribution.

‍

Introduction

Attackers have moved beyond using AI for phishing and are running agentic frameworks; open-source tools like ARTEX, CAIRN, PentAGI and Hexstrike that let AI agents scan, exploit and move through an environment with little human input [1]. ARTEX has moved quickly from an offensive security project into live intrusion activity. Since at least August 2026, the framework has been associated with espionage, financially motivated attacks and opportunistic campaigns across government, financial services, technology, defense, telecommunications and other sectors. The FBI has documented ARTEX activity spanning reconnaissance, exploitation, credential abuse, lateral movement, collection, command and control, and exfiltration [2].

Darktrace has observed activity in customer environments that overlaps with infrastructure, tooling and behaviors reported in several of these campaigns. The cases below highlight how those intrusions appeared on the network, including the activity visible before an analyst can determine which actor or framework may be responsible.

ARTEX as part of the attacker workflow

ARTEX is an open-source, multi-agent offensive security platform that converts high-level objectives into actions carried out by specialized agents [2]. Operators have used it to automate activity across the attack lifecycle, including scanning, exploitation, credential activity and post-compromise operations. Some reported intrusions moved from initial targeting to privileged access in under three hours [2].

The framework is now associated with actors pursuing different objectives and using different post-exploitation tooling. The original public repository was removed in October 2026, but archived copies and localized versions remain available. For defenders, ARTEX is more useful as context for an investigation than as an attribution shortcut.

Darktrace observations show the same point at network level. Infrastructure linked to reported ARTEX campaigns appeared alongside beaconing, tunneling, unusual TLS, reconnaissance, file retrieval and lateral activity.

Case 1: Espionage-linked activity

Darktrace observed activity associated with BLOODALCHEMY command-and-control (C2) infrastructure in two customer environments. In one environment, five devices communicated with nginxmysql[.]com and 45.153.126[.]223 between October 23 and November 2. Both indicators had previously been linked to China-nexus espionage activity involving ARTEX [4].

The connections were accompanied by suspicious beaconing, repeated successful connections and unusual self-signed TLS. One system began making suspicious TLS connections to nginxmysql[.]com on October 23. Beaconing followed on October 24, and SMB scanning appeared later in the sequence. Other affected systems subsequently communicated with previously unseen external endpoints.

Prior to the observed C2, Darktrace detected curl requests retrieving executable and script content including /systemd, /wazuh, linux_amd64, u.sh and q.sh. Related telemetry included DRSGetNCChanges activity consistent with DCSync, LDAP enumeration, DNS zone transfers, network scanning, certificate downloads and SMB writes involving start.bat.

‍

Figure 1: Darktrace investigation showing suspicious network activity associated with infrastructure linked to reported ARTEX activity.

‍

Figure 2: Darktrace detection showing anomalous self-signed TLS activity during the espionage-linked investigation.

‍

BLOODALCHEMY has been reported in more than one China-nexus cluster [5]. SoftEther infrastructure observed in one environment provided another point of correlation with reported VAPOR PANDA tradecraft. The moderate-confidence assessment tying this to a Chinese nexus threat actor is based on infrastructure, victim context, tradecraft behavior that followed the initial connection, and public reporting.

Case 2: Financial-sector activity

In early November, Darktrace observed activity in two financial-sector environments that overlap with reported ARTEX infrastructure. On November 5, a server connected to 213.111.150[.]145:8099 and downloaded wisp-agent, associated with WISP C2 activity.

Figure 3: Darktrace's Real-Time AI Analyst investigation showing a server making repeated HTTP requests, using a curl user agent, to a rare external endpoint on port 8099.

‍

In a second environment, a device attempted connections to 85.11.167[.]52 on November 1 and 2. The connections occurred alongside beaconing to new and rare destinations, anomalous TLS, possible tunneling and unusual package downloads.

The device also communicated with temp[.]sh, ssh.localhost[.]run, localhost[.]run, termbin[.]com, pastebin[.]com and GitHub-hosted release infrastructure. The sequence showed potential tooling retrieval, C2, tunneling and use of external services capable of staging or moving data.

Figure 4: Darktrace investigation showing connections to new and rare external destinations during the financial-sector activity.

‍

Case 3: Credential collection in Azure

On October 20, Darktrace observed a sweep of credential collection in a customer’s Azure environment that is consistent with ARTEX use. The intrusion began with connections to the virtual network from 43.98.196[.]73 (Alibaba Cloud). Open-source scanning data associates the address with ARTEX.

A compromised user identity made requests to the Azure Resource Manager API, though how it was compromised is unknown. The requests listed keys and secrets across API Management, web and function apps, Cognitive Services, Event Hubs, storage accounts and a managed Kubernetes cluster. With those credentials the attacker compromised multiple application identities. The attacker then used the application identities and the user identity to retrieve more credentials, change firewall and security rules and run scripts on virtual machines. Microsoft Defender alerts showed attempted execution of scripts related to Mimikatz and Impacket.

In a cloud environment, such privileged and enumerative operations are potentially indistinguishable from legitimate administrative actions. Consequently, the sequence and context of such operations signified more than any specific API call. There was no C2 traffic to match, and listing keys and using application identities are also what administrators and automation do. A compromised user, credential retrieval across several Azure services, new application identities, more credential collection and changes to security controls only made sense together, and by then access was spreading through the cloud environment.

‍

Figure 5: Darktrace Advanced Search showing Azure AD operation names associated with the activity.

‍

Figure 6: Darktrace Advanced Search showing the Azure Resource Manager action used to execute scripts.
Figure 7: Darktrace Device Event log window showing the triggered models related to the suspicious Azure activity.

‍

Shared tooling weakens attribution signals

The three cases looked different on the network: BLOODALCHEMY infrastructure and espionage-consistent tradecraft, WISP infrastructure in financial-sector systems, and Azure identity and control-plane activity with no C2 at all.

Shared tooling has complicated attribution for years, and ARTEX adds to it because the framework carries out more of the attack, so reconnaissance, exploitation attempts and some post-compromise activity can look the same whoever set the objective. Attribution still needs infrastructure, malware, victimology, timing and the sequence of behavior inside the environment, and an ARTEX link is one input to that. Moreover, the behavior and the agent will still be shaped by the biases, preferences, and tradecraft of its operators.

At its core, ARTEX’s planner and worker agents identify and record facts, while exploring intents. However, the way the Large Language Model (LLM) chooses its next steps toward an objective is shaped fundamentally by the skills and tools available in its context. This includes its starting context, as well as various possible Model Context Protocol (MCP) plugins and user-injected context (e.g. user input mid-operation). Operators still have preferred tradecraft, and this preference shapes the pathways the harness will provide to the LLM in context prompts. Even on the same platform, customization based on operator preferences and goals will remain a key facet of campaign and actor identification.

Detection and response

Across the Darktrace cases, the stronger signals came from behavior after the connection: beaconing, abnormal TLS, reconnaissance, unusual authentication, tunneling, script retrieval, SMB activity and outbound transfer. Automated activity also produced compressed bursts of requests, sequential enumeration and repeated attempts with small changes.

With agentic tools reaching privileged access in under three hours, analysts may have only a short window to connect scanning or authentication activity to credential changes, new administrative access, tooling retrieval or lateral movement. Keeping network, identity and cloud telemetry together shows what happened before and after the first match, and whether a connection was noise or an intrusion moving toward privileged access.

In some of these cases, Darktrace's Autonomous Response took or recommended action against the activity, including blocking internal connectivity over specified ports, inhibiting outbound connections with likely C2 endpoints, and/or disabling SaaS/IaaS user accounts. While many of these recommended actions required human confirmation due to deployment configuration preferences, had these actions been applied autonomously at the time of alert, the scope of activity would have been limited, thereby giving analysts time to investigate and respond.

‍

Figure 8: Autonomous Response blocking suspicious firewall modification activity.

‍

Figure 9: Autonomous Response summary for affected cloud users during the intrusion.

‍

Figure 10: Autonomous Response taking action against unusual lateral movement activity on an affected device.

‍

Figure 12: Autonomous Response summary for affected network users during the time of the intrusion.

ARTEX won’t be the last of these frameworks. The next ones will come with new capabilities and defaults, but an intrusion still must interact with the target environment, and that shows up in how systems, identities and applications behave even after the indicator has aged out. The harder intrusions to catch will have no infrastructure to match, just ordinary API calls and legitimate identities. Detecting those depends on knowing what normal looks like for each identity and device. Containing abnormal activity at machine speed can stop an intrusion progressing while analysts establish impact and its extent. Autonomous Response does not replace that human analysis, but rather augments defender teams' capabilities by extending the window for investigation.

‍

Credit to Nathaniel Jones (SVP, Global Threat Intelligence), Eugene Chua (Principal Analyst), Sam Lister (Senior Security Researcher), Adam Potter (Associate Principal Cyber Analyst), Daniel Levy (Threat Hunting Data Scientist)

Edited by Ryan Traill (Content Manager)

Appendices

Darktrace model detections

·       Anomalous Connection / Multiple HTTP POSTs to Rare Hostname

·       Anomalous Connection / Rare External SSL Self-Signed

·       Anomalous Connection / Suspicious Self-Signed SSL

·       Compromise / Large Number of Suspicious Successful Connections

·       Compromise / Suspicious Beaconing Behaviour

·       Compromise / Agent Beacon to New Endpoint

·       Compromise / Beacon to Young Endpoint

·       Compromise / SSL Beaconing to Rare Destination

·       Compromise / Suspicious TLS Beaconing To Rare External

·       Compromise / Possible Tunnelling to Bin Services

·       Device / Anomalous Github Download

·       Device / Anomalous Package Download

·       Device / Initial Attack Chain Activity

Indicators of compromise (IoCs)

nginxmysql[.]com

45.153.126[.]223

213.111.150[.]145

wisp-agent

85.11.167[.]52

43.98.196[.]73

192.121.47[.]20

www[.]ddduuuppp[.]com

89.46.234[.]50

sqlserver-sqlite[.]com

199.19.95[.]66

graphics-hook-filter64.dll

graphics-hook_config.bin

References

1.      Google. GTIG AI Threat Tracker - https://cloud.google.com/blog/topics/threat-intelligence/agentic-threat-activity-crosses-a-significant-threshold

2.      Federal Bureau of Investigation, Threat Overview of ARTEX Autonomous Penetration Testing Framework, FLASH-20261113-001, 13 November 2026: https://www.ic3.gov/CSA/2026/261113.pdf

3.      Microsoft Security Blog, Storm-0940 Uses ARTEX Agentic AI in ConnectWise Supply Chain Compromise: https://www.microsoft.com/en-us/security/blog/2026/10/21/storm-0940-uses-artex-agentic-ai-connectwise-supply-chain-compromise/

4.      CrowdStrike, VAPOR PANDA Uses ARTEX to Scale Global Campaign: https://www.crowdstrike.com/en-us/blog/vapor-panda-uses-artex-to-scale-global-campaign/

5.      Bitdefender, SilkParasite: Tracking a China-Nexus APT Across Central Asia: https://www.bitdefender.com/en-us/blog/businessinsights/silkparasite-tracking-china-nexus-apt-across-central-asia

‍

‍

Continue reading
About the author
Nathaniel Jones
SVP, Global Threat Intelligence
Your data. Our AI.
Elevate your network security with Darktrace AI