Blog
/
/
April 26, 2020

How Cyber-Criminals Leverage AI in Attacks

Cyber attacks are relentless and ever-evolving. Learn how cyber-criminals are using AI to augment their attacks at every stage of the kill chain.
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
Max Heinemeyer
Global Field CISO
Default blog image
26
Apr 2020

Overview

The mind of an experienced and dedicated cyber-criminal works like that of an entrepreneur: the relentless pursuit of profit guides every move they make. At each step of the journey towards their objective, the same questions are asked: how can I minimize my time and resources? How can I mitigate against risk? What measures can I take which will return the best results?

Incorporating this ‘enterprise’ model into the cyber-criminal framework uncovers why attackers are turning to new technology in an attempt to maximize efficiency, and why a report from Forrester earlier this year revealed that 88% of security leaders now consider the nefarious use of AI in cyber activity to be inevitable. Over half of the responders to that same survey foresee AI attacks manifesting themselves to the public in the next twelve months – or think they are already occurring.

AI has already achieved breakthroughs in fields such as healthcare, facial recognition, voice assistance and many others. In the current cat-and-mouse game of cyber security, defenders have started to accept that augmenting their defenses with AI is necessary, with over 3,500 organizations using machine learning to protect their digital environments. But we have to be ready for the moment attackers themselves use open-source AI technology available today to supercharge their attacks.

Enhancing the attack life cycle

To a cyber-criminal ring, the benefits of leveraging AI in their attacks are at least four-fold:

  • It gives them an understanding of context
  • It helps to scale up operations
  • It makes attribution and detection harder
  • It ultimately increases their profitability

To best demonstrate how each of these factors surface themselves, we can break down the life cycle of a typical data exfiltration attempt, telling the story of how AI can augment the attacker during the campaign at every stage of the attack.

ReconnaissanceCAPTCHA breakerIntrusionShellphish and SNAP_RC2 establishmentFirstOrder and unsupervised clustering algorithmPrivilege escalationCeWL and neural networkLateral movementMITRE CALDERAMission accomplishedYahoo NSFW

Figure 1: The ‘AI toolbox’ attackers use to augment their attacks

Stage 1: Reconnaissance

In seeking to garner trust and make inroads into an organization, automated chatbots would first interact with employees via social media, leveraging profile pictures of non-existent people created by AI instead of re-using actual human photos. Once the chatbots have gained the trust of the victims at the target organization, the human attackers can gain valuable intelligence about its employees, while CAPTCHA-breakers are used for automated reconnaissance on the organization’s public-facing web pages.

Forrester estimates that AI-enabled ‘deep fakes’ will cost businesses a quarter of a billion dollars in losses in 2020.

Stage 2: Intrusion

This intelligence would then be used to craft convincing spear phishing attacks, whilst an adapted version of SNAP_R can be leveraged to create realistic tweets at scale – targeting several key employees. The tweets either trick the user into downloading malicious documents, or contain links to servers which facilitate exploit-kit attacks.

An autonomous vulnerability fuzzing engine based on Shellphish would be constantly crawling the victim’s perimeter – internet-facing servers and websites – and trying to find new vulnerabilities for an initial foothold.

Stage 3: Command and control

A popular hacking framework, Empire, allows attackers to ‘blend in’ with regular business operations, restricting command and control traffic to periods of peak activity. An agent using some form of automated decision-making engine for lateral movement might not even require command and control traffic to move laterally. Eliminating the need for command and control traffic drastically reduces the detection surface of existing malware.

Stage 4: Privilege escalation

At this stage, a password crawler like CeWL could collect target-specific keywords from internal websites and feed those keywords into a pre-trained neural network, essentially creating hundreds of realistic permutations of contextualized passwords at machine-speed. These can be automatically entered in period bursts so as to not alert the security team or trigger resets.

Stage 5: Lateral movement

Moving laterally and harvesting accounts and credentials involves identifying the optimal paths to accomplish the mission and minimize intrusion time. Parts of the attack planning can be accelerated by concepts such as from the CALDERA framework using automated planning AI methods. This would greatly reduce the time required to reach the final destination.

Stage 6: Data exfiltration

It is in this final stage where the role of offensive AI is most apparent. Instead of running a costly post-intrusion analysis operation and sifting through gigabytes of data, the attackers can leverage a neural network that pre-selects only relevant material for exfiltration. This neural network is pre-trained and therefore has a basic understanding of what valuable material constitutes and flags those for immediate exfiltration. The neural network could be based on something like Yahoo’s open-source project for content recognition.

Conclusion

Today’s attacks still require several humans behind the keyboard making guesses about the sorts of methods that will be most effective in their target network – it’s this human element that often allows defenders to neutralize attacks.

Offensive AI will make detecting and responding to attacks far more difficult. Open-source research and projects exist today which can be leveraged to augment every phase of the attack lifecycle. This means that the speed, scale, and contextualization of attacks will exponentially increase. Traditional security controls are already struggling to detect attacks that have never been seen before in the wild – be it malware without known signatures, new command and control domains, or individualized spear phishing emails. There is no chance that traditional tools will be able to cope with future attacks as this becomes the norm and easier to realize than ever before.

To stay ahead of this next wave of attacks, AI is becoming a necessary part of the defender’s stack, as no matter how well-trained or how well-staffed, humans alone will no longer be able to keep up. Hundreds of organizations are already using Autonomous Response to fight back against new strains of ransomware, insider threats, previously unknown techniques, tools and procedures, and many other threats. Cyber AI technology allows human responders to take stock and strategize from behind the front line. A new age in cyber defense is just beginning, and the effect of AI on this battleground is already proving fundamental.

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
Max Heinemeyer
Global Field CISO

More in this series

No items found.

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

Blog

/

AI

/

November 18, 2026

Darktrace qualifies for NATO’s Cybersecurity Dynamic Marketplace

Default blog imageDefault blog image

Darktrace has been selected to join NATO’s Cybersecurity Dynamic Marketplace (CSDM), an initiative designed to give NATO faster, more flexible access to cybersecurity technology and services in support of its missions.  

Established by the NATO Communications and Information Agency (NCIA), the marketplace brings together qualified suppliers whose technical capabilities have been reviewed, helping NATO match specialist expertise to project needs and respond more quickly to changing operational priorities.

Qualified across all eight cybersecurity categories

The program’s first phase, known as Lot 1, covers eight cybersecurity categories. Suppliers can qualify in individual categories; Darktrace has qualified to compete for future work across all eight:

  • Identity, Credential and Access Management
  • Endpoint Protection and Platforms
  • Network and Perimeter Security
  • Cloud Security
  • Threat Detection and Analytics Platforms
  • Vulnerability and Exposure Management
  • Digital Forensics and Evidence Platforms
  • Professional Services

Darktrace’s qualification across all eight categories reflects the breadth of our technology and services.

Supporting government and defense organizations

Nation-state attacks, compromised suppliers, the rapid exploitation of security weaknesses and attackers’ growing use of AI are increasing pressure on public-sector organizations. At the same time, agencies are adopting cloud services, connecting the technology that runs physical operations, and introducing automation and AI. Security teams must manage these changes while keeping essential operations running and sensitive information secure.

Protecting these environments requires an understanding of how people, AI and connected systems interact, and the ability to recognize when their activity creates risk. Threats can span several systems, making it essential for security teams to connect related activity and respond quickly.

The Darktrace Behavioral Defense Platform™ supports this approach through unified visibility, continuous behavioral monitoring and autonomous response across AI, people and infrastructure.

At its core, Darktrace’s Adaptive AI™ continuously learns how each organization behaves, building a unique understanding of its people, AI agents and connected systems. This enables the platform to help identify activity that does not belong, connect related behavior across the organization, and investigate and respond to threats autonomously in real time, helping prevent disruption to essential operations.

Our selection builds on work to meet government security requirements internationally. Darktrace recently completed an independent assessment under Australia’s InfoSec Registered Assessors Program (IRAP) for Darktrace / NETWORK™ and Darktrace / OT™ against the Australian Government’s Information Security Manual PROTECTED-level controls. In the United States, Darktrace Federal continues to deliver the Darktrace technology to Department of War, Department of State, and Federal civilian organizations via its on-premises and FedRAMP CLASS D (High) Authorization platforms.

Together, these achievements demonstrate Darktrace’s ability to meet rigorous public-sector security requirements and reinforce its position as a trusted provider of mission-critical cybersecurity technology for government environments.

Meet us at NATO Edge 2026

Darktrace representatives will attend NATO Edge 2026, NCIA’s flagship technology event, in İzmir, Türkiye, from 17 to 19 November. We look forward to meeting government, military, and industry partners to discuss their cybersecurity priorities and how we can support them.

To learn more about the Darktrace Behavioral Defense Platform, visit the platform hub.

Continue reading
About the author
The Darktrace Community
Your data. Our AI.
Elevate your network security with Darktrace AI