Blog
/
OT
/
February 9, 2022

The Impact of Conti Ransomware on OT Systems

Learn how ransomware can spread throughout converged IT/OT environments, and how Self-Learning AI empowers organizations to contain these threats.
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
Oakley Cox
Director of Product
Default blog imageDefault blog imageDefault blog imageDefault blog imageDefault blog imageDefault blog image
09
Feb 2022

Ransomware has taken the world by storm, and IT is not the only technology affected. Operational Technology (OT), which is increasingly blending with IT, is also susceptible to ransomware tactics, techniques, and procedures (TTPs). And when ransomware strikes OT, the effects have the potential to be devastating.

Here, we will look at a ransomware attack that spread from IT to OT systems. The attack was detected by Darktrace AI.

This threat find demonstrates a use case of Darktrace’s technology that delivers immense value to organizations with OT: spotting and stopping ransomware at its earliest stages, before the damage is done. This is particularly helpful for organizations with interconnected enterprise and industrial environments, as it means:

  1. Emerging attacks can be contained in IT before they spread laterally into OT, and even before they spread from device to device in IT;
  2. Organizations gain granular visibility into their industrial environments, detecting deviations from normal activity, and quick identification of remediating actions.

Threat find: Ransomware and crypto-mining hijack affecting IT and OT systems

Darktrace recently identified an aggressive attack targeting an OT R&D investment firm in EMEA. The attack originally started as a crypto-mining campaign and later evolved into ransomware. This organization deployed Darktrace in a digital estate containing both IT and OT assets that spanned over 3,000 devices.

If the organization had deployed Darktrace’s Autonomous Response technology in active mode, this threat would have been stopped in its earliest stages. Even in the absence of Autonomous Response, however, mere human attention would have stopped this attack’s progression. Darktrace’s Self-Learning AI gave clear indications of an ongoing compromise in the month prior to the detonation of ransomware. In this case, however, the security team was not monitoring Darktrace’s interface, and so the attack was allowed to proceed.

Compromised OT devices

This threat find will focus on the attack techniques used to take over two OT devices, specifically, a HMI (human machine interface), and an ICS Historian used to collect and log industrial data. These OT devices were both VMware virtual machines running Windows OS, and were compromised as part of a wider Conti ransomware infection. Both devices were being used primarily within an industrial control system (ICS), running a popular ICS software package and making regular connections to an industrial cloud platform.

These devices were thus part of an ICSaaS (ICS-as-a-Service) environment, using virtualised and Cloud platforms to run analytics, update threat intelligence, and control the industrial process. As previously highlighted by Darktrace, the convergence of cloud and ICS increases a network’s attack surface and amplifies cyber risk.

Attack lifecycle

Opening stages

The initial infection of the OT devices occurred when a compromised Domain Controller (DC) made unusual Active Directory requests. The devices made subsequent DCE-RPC binds for epmapper, often used by attackers for command execution, and lsarpc, used by attackers to abuse authentication policies and escalate privileges.

The payload was delivered when the OT devices used SMB to connect to the sysvol folder on the DC and read a malicious executable file, called SetupPrep.exe.

Figure 1: Darktrace model breaches across the whole network from initial infection on October 21 to the detonation on November 15.

Figure 2: ICS reads on the HMI in the lead up, during, and following detonation of the ransomware.

Device encryption and lateral spread

The malicious payload remained dormant on the OT devices for three weeks. It seems the attacker used the time to install crypto-mining malware elsewhere on the network and consolidate their foothold.

On the day the ransomware detonated, the attacker used remote management tools to initiate encryption. The PSEXEC tool was used on an infected server (separate from the original DC) to remotely execute malicious .dll files on the compromised OT devices.

The devices then attempted to make command and control (C2) connections to rare external endpoints using suspicious ports. Like in many ICS networks, sufficient network segregation had been implemented to prevent the HMI device from making successful connections to the Internet and the C2 communications failed. But worryingly, the failed C2 did not prevent the attack from proceeding or the ransomware from detonating.

The Historian device made successful C2 connections to around 40 unique external endpoints. Darktrace detected beaconing-type behavior over suspicious TCP/SSL ports including 465, 995, 2078, and 2222. The connections were made to rare destination IP addresses that did not specify the Server Name Indication (SNI) extension hostname and used self-signed and/or expired SSL certificates.

Both devices enumerated network SMB shares and wrote suspicious shell scripts to network servers. Finally, the devices used SMB to encrypt files stored in network shares, adding a file extension which is likely to be unique to this victim and which will be called ABCXX for the purpose of this blog. Most encrypted files were uploaded to the folder in which the file was originally located, but in some instances were moved to the images folder.

During the encryption, the device was using the machine account to authenticate SMB sessions. This is in contrast to other ransomware incidents that Darktrace has observed, in which admin or service accounts are compromised and abused by the attacker. It is possible that in this instance the attacker was able to use ‘Living off the Land’ techniques (for example the use of lsarpc pipe) to give the machine account admin privileges.

Examples of files being encrypted and moved:

  • SMB move success
  • File: new\spbr0007\0000006A.bak
  • Renamed: new\spbr0007\0000006A.bak.ABCXX
  • SMB move success
  • File: ActiveMQ\readme.txt
  • Renamed: Images\10j0076kS1UA8U975GC2e6IY.488431411265952821382.png.ABCXX

Detonation of ransomware

Upon detonation, the ransomware note readme.txt was written by the ICS to targeted devices as part of the encryption activity.

The final model breached by the device was “Unresponsive ICS Device” as the device either stopped working due to the effects of the ransomware, or was removed from the network.

Figure 3: abc-histdev — external connections filtered on destination port 995 shows C2 connections starting around one hour before encryption began.

How the attack bypassed the rest of the security stack

In this threat find, there were a number of factors which resulted in the OT devices becoming compromised.

The first is IT/OT convergence. The ICS network was insufficiently segregated from the corporate network. This means that devices could be accessed by the compromised DC during the lateral movement stage of the attack. As OT becomes more reliant on IT, ensuring sufficient segregation is in place, or that an attacker can not circumvent such segregation, is becoming an ever increasing challenge for security teams.

Another reason is that the attacker used attack methods which leverage Living off the Land techniques to compromise devices with no discrimination as to whether they were part of an IT or OT network. Many of the machines used to operate ICS networks, including the devices highlighted here, rely on operating systems vulnerable to the kinds of TTPs observed here and that are regularly employed by ransomware groups.

Darktrace insights

Darktrace’s Cyber AI Analyst was able to stitch together many disparate forms of unusual activity across the compromised devices to give a clear security narrative containing details of the attack. The incident report for the Historian server is shown below. This provides a clear illustration of how Cyber AI Analyst can close any skills or communication gap between IT and OT specialists.

Figure 4: Cyber AI Analyst of the Historian server (abc-histdev). It investigated and reported the C2 communication (step 2) that started just before network reconnaissance using TCP scanning (step 3) and the subsequent file encryption over SMB (step 4).

In total, the attacker’s dwell time within the digital estate was 25 days. Unfortunately, it lead to disruption to operational technology, file encryption and financial loss. Altogether, 36 devices were crypto-mining for over 20 days – followed by nearly 100 devices (IT and OT) becoming encrypted following the detonation of the ransomware.

If it were active, Autonomous Response would have neutralized this activity, containing the damage before it could escalate into crisis. Darktrace’s Self-Learning AI gave clear indications of an ongoing compromise in the month prior to the detonation of ransomware, and so any degree of human attention toward Darktrace’s revelations would have stopped the attack.

Autonomous Response is highly configurable, and so, in industrial environments — whether air-gapped OT or converged IT/OT ecosystems — Antigena can be deployed in a variety of manners. In human confirmation mode, human operators need to give the green light before the AI takes action. Antigena can also be deployed only in the higher levels of the Purdue model, or the “IT in OT,” protecting the core assets from fast-moving attacks like ransomware.

Ransomware and interconnected IT/OT systems

ICS networks are often operated by machines that rely on operating systems which can be affected by TTPs regularly employed by ransomware groups — that is, TTPs such as Living off the Land, which do not discriminate between IT and OT.

The threat that ransomware poses to organizations with OT, including critical infrastructure, is so severe that the Cyber Infrastructure and Security Agency (CISA) released a fact sheet concerning these threats in the summer of 2021, noting the risk that IT attacks pose to OT networks:

“OT components are often connected to information technology (IT) networks, providing a path for cyber actors to pivot from IT to OT networks… As demonstrated by recent cyber incidents, intrusions affecting IT networks can also affect critical operational processes even if the intrusion does not directly impact an OT network.”

Major ransomware attacks against the Colonial Pipeline and JBS Foods demonstrate the potential for ransomware affecting OT to cause severe economic disruption on a national and international scale. And ransomware can wreak havoc on OT systems regardless of whether they directly target OT systems.

As industrial environments continue to converge and evolve — be they IT/OT, ICSaaS, or simply poorly segregated legacy systems — Darktrace stands ready to contain attacks before the damage is done. It is time for organizations with industrial environments to take the quantum leap forward that Darktrace’s Self-Learning AI is uniquely positioned to provide.

Thanks to Darktrace analysts Ash Brice and Andras Balogh for their insights on the above threat find.

Discover more on how Darktrace protects OT environments from ransomware

Darktrace model detections

HMI in chronological order at time of detonation:

  • Anomalous Connection / SMB Enumeration
  • Anomalous File / Internal / Unusual SMB Script Write
  • Anomalous File / Internal / Additional Extension Appended to SMB File
  • Compromise / Ransomware / Suspicious SMB Activity [Enhanced Monitoring]
  • ICS / Unusual Data Transfer By OT Device
  • ICS / Unusual Unresponsive ICS Device

Historian

  • ICS / Rare External from OT Device
  • Anomalous Connection / Anomalous SSL without SNI to New External
  • Anomalous Connection / Multiple Connections to New External TCP Port
  • ICS / Unusual Activity From OT Device
  • Anomalous Connection / SMB Enumeration
  • Anomalous Connection / Suspicious Activity On High Risk Device
  • Unusual Activity / SMB Access Failures
  • Device / Large Number of Model Breaches
  • ICS / Unusual Data Transfer By OT Device
  • Anomalous File / Internal / Additional Extension Appended to SMB File
  • Device / SMB Lateral Movement
  • Compromise / Ransomware / Suspicious SMB Activity [Enhanced Monitoring]
  • Device / Multiple Lateral Movement Model Breaches [Enhanced Monitoring]

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
Oakley Cox
Director of Product

More in this series

No items found.

Blog

/

/

April 16, 2025

Introducing Version 2 of Darktrace’s Embedding Model for Investigation of Security Threats (DEMIST-2)

woman looking at laptop at deskDefault blog imageDefault blog image

DEMIST-2 is Darktrace’s latest embedding model, built to interpret and classify security data with precision. It performs highly specialized tasks and can be deployed in any environment. Unlike generative language models, DEMIST-2 focuses on providing reliable, high-accuracy detections for critical security use cases.

DEMIST-2 Core Capabilities:  

  • Enhances Cyber AI Analyst’s ability to triage and reason about security incidents by providing expert representation and classification of security data, and as a part of our broader multi-layered AI system
  • Classifies and interprets security data, in contrast to language models that generate unpredictable open-ended text responses  
  • Incorporates new innovations in language model development and architecture, optimized specifically for cybersecurity applications
  • Deployable across cloud, on-prem, and edge environments, DEMIST-2 delivers low-latency, high-accuracy results wherever it runs. It enables inference anywhere.

Cybersecurity is constantly evolving, but the need to build precise and reliable detections remains constant in the face of new and emerging threats. Darktrace’s Embedding Model for Investigation of Security Threats (DEMIST-2) addresses these critical needs and is designed to create stable, high-fidelity representations of security data while also serving as a powerful classifier. For security teams, this means faster, more accurate threat detection with reduced manual investigation. DEMIST-2's efficiency also reduces the need to invest in massive computational resources, enabling effective protection at scale without added complexity.  

As an embedding language model, DEMIST-2 classifies and creates meaning out of complex security data. This equips our Self-Learning AI with the insights to compare, correlate, and reason with consistency and precision. Classifications and embeddings power core capabilities across our products where accuracy is not optional, as a part of our multi-layered approach to AI architecture.

Perhaps most importantly, DEMIST-2 features a compact architecture that delivers analyst-level insights while meeting diverse deployment needs across cloud, on-prem, and edge environments. Trained on a mixture of general and domain-specific data and designed to support task specialization, DEMIST-2 provides privacy-preserving inference anywhere, while outperforming larger general-purpose models in key cybersecurity tasks.

This proprietary language model reflects Darktrace's ongoing commitment to continually innovate our AI solutions to meet the unique challenges of the security industry. We approach AI differently, integrating diverse insights to solve complex cybersecurity problems. DEMIST-2 shows that a refined, optimized, domain-specific language model can deliver outsized results in an efficient package. We are redefining possibilities for cybersecurity, but our methods transfer readily to other domains. We are eager to share our findings to accelerate innovation in the field.  

The evolution of DEMIST-2

Key concepts:  

  • Tokens: The smallest units processed by language models. Text is split into fragments based on frequency patterns allowing models to handle unfamiliar words efficiently
  • Low-Rank Adaptors (LoRA): Small, trainable components added to a model that allow it to specialize in new tasks without retraining the full system. These components learn task-specific behavior while the original foundation model remains unchanged. This approach enables multiple specializations to coexist, and work simultaneously, without drastically increasing processing and memory requirements.

Darktrace began using large language models in our products in 2022. DEMIST-2 reflects significant advancements in our continuous experimentation and adoption of innovations in the field to address the unique needs of the security industry.  

It is important to note that Darktrace uses a range of language models throughout its products, but each one is chosen for the task at hand. Many others in the artificial intelligence (AI) industry are focused on broad application of large language models (LLMs) for open-ended text generation tasks. Our research shows that using LLMs for classification and embedding offers better, more reliable, results for core security use cases. We’ve found that using LLMs for open-ended outputs can introduce uncertainty through inaccurate and unreliable responses, which is detrimental for environments where precision matters. Generative AI should not be applied to use cases, such as investigation and threat detection, where the results can deeply matter. Thoughtful application of generative AI capabilities, such as drafting decoy phishing emails or crafting non-consequential summaries are helpful but still require careful oversight.

Data is perhaps the most important factor for building language models. The data used to train DEMIST-2 balanced the need for general language understanding with security expertise. We used both publicly available and proprietary datasets.  Our proprietary dataset included privacy-preserving data such as URIs observed in customer alerts, anonymized at source to remove PII and gathered via the Call Home and aianalyst.darktrace.com services. For additional details, read our Technical Paper.  

DEMIST-2 is our way of addressing the unique challenges posed by security data. It recognizes that security data follows its own patterns that are distinct from natural language. For example, hostnames, HTTP headers, and certificate fields often appear in predictable ways, but not necessarily in a way that mirrors natural language. General-purpose LLMs tend to break down when used in these types of highly specialized domains. They struggle to interpret structure and context, fragmenting important patterns during tokenization in ways that can have a negative impact on performance.  

DEMIST-2 was built to understand the language and structure of security data using a custom tokenizer built around a security-specific vocabulary of over 16,000 words. This tokenizer allows the model to process inputs more accurately like encoded payloads, file paths, subdomain chains, and command-line arguments. These types of data are often misinterpreted by general-purpose models.  

When the tokenizer encounters unfamiliar or irregular input, it breaks the data into smaller pieces so it can still be processed. The ability to fall back to individual bytes is critical in cybersecurity contexts where novel or obfuscated content is common. This approach combines precision with flexibility, supporting specialized understanding with resilience in the face of unpredictable data.  

Along with our custom tokenizer, we made changes to support task specialization without increasing model size. To do this, DEMIST-2 uses LoRA . LoRA is a technique that integrates lightweight components with the base model to allow it to perform specific tasks while keeping memory requirements low. By using LoRA, our proprietary representation of security knowledge can be shared and reused as a starting point for more highly specialized models, for example, it takes a different type of specialization to understand hostnames versus to understand sensitive filenames. DEMIST-2 dynamically adapts to these needs and performs them with purpose.  

The result is that DEMIST-2 is like having a room of specialists working on difficult problems together, while sharing a basic core set of knowledge that does not need to be repeated or reintroduced to every situation. Sharing a consistent base model also improves its maintainability and allows efficient deployment across diverse environments without compromising speed or accuracy.  

Tokenization and task specialization represent only a portion of the updates we have made to our embedding model. In conjunction with the changes described above, DEMIST-2 integrates several updated modeling techniques that reduce latency and improve detections. To learn more about these details, our training data and methods, and a full write-up of our results, please read our scientific whitepaper.

DEMIST-2 in action

In this section, we highlight DEMIST-2's embeddings and performance. First, we show a visualization of how DEMIST-2 classifies and interprets hostnames, and second, we present its performance in a hostname classification task in comparison to other language models.  

Embeddings can often feel abstract, so let’s make them real. Figure 1 below is a 2D visualization of how DEMIST-2 classifies and understands hostnames. In reality, these hostnames exist across many more dimensions, capturing details like their relationships with other hostnames, usage patterns, and contextual data. The colors and positions in the diagram represent a simplified view of how DEMIST-2 organizes and interprets these hostnames, providing insights into their meaning and connections. Just like an experienced human analyst can quickly identify and group hostnames based on patterns and context, DEMIST-2 does the same at scale.  

DEMIST-2 visualization of hostname relationships from a large web dataset.
Figure 1: DEMIST-2 visualization of hostname relationships from a large web dataset.

Next, let’s zoom in on two distinct clusters that DEMIST-2 recognizes. One cluster represents small businesses (Figure 2) and the other, Russian and Polish sites with similar numerical formats (Figure 3). These clusters demonstrate how DEMIST-2 can identify specific groupings based on real-world attributes such as regional patterns in website structures, common formats used by small businesses, and other properties such as its understanding of how websites relate to each other on the internet.

Cluster of small businesses
Figure 2: Cluster of small businesses
Figure 3: Cluster of Russian and Polish sites with a similar numerical format

The previous figures provided a view of how DEMIST-2 works. Figure 4 highlights DEMIST-2’s performance in a security-related classification task. The chart shows how DEMIST-2, with just 95 million parameters, achieves nearly 94% accuracy—making it the highest-performing model in the chart, despite being the smallest. In comparison, the larger model with 2.78 billion parameters achieves only about 89% accuracy, showing that size doesn’t always mean better performance. Small models don’t mean poor performance. For many security-related tasks, DEMIST-2 outperforms much larger models.

Hostname classification task performance comparison against comparable open source foundation models
Figure 4: Hostname classification task performance comparison against comparable open source foundation models

With these examples of DEMIST-2 in action, we’ve shown how it excels in embedding and classifying security data while delivering high performance on specialized security tasks.  

The DEMIST-2 advantage

DEMIST-2 was built for precision and reliability. Our primary goal was to create a high-performance model capable of tackling complex cybersecurity tasks. Optimizing for efficiency and scalability came second, but it is a natural outcome of our commitment to building a strong, effective solution that is available to security teams working across diverse environments. It is an enormous benefit that DEMIST-2 is orders of magnitude smaller than many general-purpose models. However, and much more importantly, it significantly outperforms models in its capabilities and accuracy on security tasks.  

Finding a product that fits into an environment’s unique constraints used to mean that some teams had to settle for less powerful or less performant products. With DEMIST-2, data can remain local to the environment, is entirely separate from the data of other customers, and can even operate in environments without network connectivity. The size of our model allows for flexible deployment options while at the same time providing measurable performance advantages for security-related tasks.  

As security threats continue to evolve, we believe that purpose-built AI systems like DEMIST-2 will be essential tools for defenders, combining the power of modern language modeling with the specificity and reliability that builds trust and partnership between security practitioners and AI systems.

Conclusion

DEMIST-2 has additional architectural and deployment updates that improve performance and stability. These innovations contribute to our ability to minimize model size and memory constraints and reflect our dedication to meeting the data handling and privacy needs of security environments. In addition, these choices reflect our dedication to responsible AI practices.

DEMIST-2 is available in Darktrace 6.3, along with a new DIGEST model that uses GNNs and RNNs to score and prioritize threats with expert-level precision.

[related-resource]

Continue reading
About the author
Margaret Cunningham, PhD
Director, Security & AI Strategy, Field CISO

Blog

/

/

April 16, 2025

AI Uncovered: Introducing Darktrace Incident Graph Evaluation for Security Threats (DIGEST)

man looking at computer screenDefault blog imageDefault blog image

DIGEST advances how Cyber AI Analyst scores and prioritizes incidents. Trained on over a million anonymized incident graphs, our model brings deeper context to severity scoring by analyzing how threats are structured and how they evolve. DIGEST assesses threats as an expert, before damage is done. For more details beyond this overview, please read our Technical Research Paper.

Darktrace combines machine learning (ML) and artificial intelligence (AI) approaches using a multi-layered, multi-method approach. The result is an AI system that continuously ingests data from across an organization’s environment, learns from it, and adapts in real time. DIGEST adds a new layer to this system, specifically to our Cyber AI Analyst, the first and most experienced AI Analyst in cybersecurity, dedicated to refining how incidents are scored and prioritized. DIGEST improves what your team uses to focus on what matters the most first.

To build DIGEST, we combined Graph Neural Networks (GNNs) to interpret incident structure with Recurrent Neural Networks (RNNs) to analyze how incidents evolve over time. This pairing allows DIGEST to reliably determine the potential severity of an incident even at an early stage to give the Cyber AI Analyst a critical edge in identifying high-risk threats early and recognizing when activity is unlikely to escalate.

DIGEST works locally in real-time regardless of whether your Darktrace deployment is on prem or in the cloud, without requiring data to be sent externally for decisions to be made. It was built to support teams in all environments, including those with strict data controls and limited connectivity.

Our approach to AI is unique, drawing inspiration from multiple disciplines to tackle the toughest cybersecurity challenges. DIGEST demonstrates how a novel application of GNNs and RNNs improves the prioritization and triage of security incidents. By blending interdisciplinary expertise with innovative AI techniques, we are able to push the boundaries of what’s possible and deliver it where it is needed most. We are eager to share our findings to accelerate progress throughout the broader field of AI development.

DIGEST: Pattern, progression, and prioritization

Most security incidents start quietly. A device contacting an unusual domain. Credentials are used at unexpected hours. File access patterns shift. The fundamental challenge is not always detecting these anomalies but knowing what to address first. DIGEST gives us this capability.

To understand DIGEST, it helps to start with Cyber AI Analyst, a critical component of our Self-Learning AI system and a front-line triage partner in security investigations. It combines supervised and unsupervised machine learning (ML) techniques, natural language processing (NLP), and graph-based reasoning to investigate and summarize security incidents.

DIGEST was built as an additional layer of analysis within Cyber AI Analyst. It enhances its capabilities by refining how incidents are scored and prioritized, helping teams focus on what matters most more quickly. For a general view of the ML and AI methods that power Darktrace products, read our AI Arsenal whitepaper. This paper provides insights regarding the various approaches we use to detect, investigate, and prioritize threats.

Cyber AI Analyst is constantly investigating alerts and produces millions of critical incidents every year. The dynamic graphs produced by Cyber AI Analyst investigations represent an abstract understanding of security incidents that is fully anonymized and privacy preserving. This allowed us to use the Call Home and aianalyst.darktrace.com services to produce a dataset comprising the broad structure of millions of incidents that Cyber AI analyst detected on customer deployments, without containing any sensitive data. (Read our technical research paper for more details about our dataset).

The dynamic graphs from Cyber AI Analyst capture the structure of security incidents where nodes represent entities like users, devices or resources, and edges represent the multitude of relationships between them. As new activity is observed, the graph expands, capturing the progression of incidents over time. Our dataset contained everything from benign administrative behavior to full-scale ransomware attacks.

Unique data, unmatched insights

Key terms

Graph Neural Networks (GNNs): A type of neural network designed to analyze and interpret data structured as graphs, capturing relationships between nodes.

Recurrent Neural Networks (RNNs): A type of neural network designed to model sequences where the order of events matters, like how activity unfolds in a security incident.

The Cyber AI Analyst dataset used to train DIGEST reflects over a decade of work in AI paired with unmatched expertise in cybersecurity. Prior to training DIGEST on our incident graph data set, we performed rigorous data preprocessing to ensure to remove issues such as duplicate or ill-formed incidents. Additionally, to validate DIGEST’s outputs, expert security analysts assessed and verified the model’s scoring.

Transforming data into insights requires using the right strategies and techniques. Given the graphical nature of Cyber AI Analyst incident data, we used GNNs and RNNs to train DIGEST to understand incidents and how they are likely to change over time. Change does not always mean escalation. DIGEST’s enhanced scoring also keeps potentially legitimate or low-severity activity from being prioritized over threats that are more likely to get worse. At the beginning, all incidents might look the same to a person. To DIGEST, it looks like the beginning of a pattern.

As a result, DIGEST enhances our understanding of security incidents by evaluating the structure of the incident, probable next steps in an incident’s trajectory, and how likely it is to grow into a larger event.

To illustrate these capabilities in action, we are sharing two examples of DIGEST’s scoring adjustments from use cases within our customers’ environments.

First, Figure 1 shows the graphical representation of a ransomware attack, and Figure 2 shows how DIGEST scored incident progression of that ransomware attack. At hour two, DIGEST’s score escalated to 95% well before observation of data encryption. This means that prior to seeing malicious encryption behaviors, DIGEST understood the structure of the incident and flagged these early activities as high-likelihood precursors to a severe event. Early detection, especially when flagged prior to malicious encryption behaviors, gives security teams a valuable head start and can minimize the overall impact of the threat, Darktrace Autonomous Response can also be enabled by Cyber AI Analyst to initiate an immediate action to stop the progression, allowing the human security team time to investigate and implement next steps.

Graph representation of a ransomware attack
Figure 1: Graph representation of a ransomware attack
Timeline of DIGEST incident score escalation. Note that timestep does not equate to hours, the spike in score to 95% occurred approximately 2 hours into the attack, prior to data encryption.
Figure 2:  Timeline of DIGEST incident score escalation. Note that timestep does not equate to hours, the spike in score to 95% occurred approximately 2 hours into the attack, prior to data encryption.

In contrast, our second example shown in Figure 3 and Figure 4 illustrates how DIGEST’s analysis of an incident can help teams avoid wasting time on lower risk scenarios. In this instance, Figure 3 illustrates a graph of unusual administrative activity, where we observed connection to a large group of devices. However, the incident score remained low because DIGEST determined that high risk malicious activity was unlikely. This determination was based on what DIGEST observed in the incident's structure, what it assessed as the probable next steps in the incident lifecycle and how likely it was to grow into a larger adverse event.

Graph representation of unusual admin activity connecting to a large group of devices.
Figure 3: Graph representation of unusual admin activity connecting to a large group of devices.
Timeline of DIGEST incident scoring, where the score remained low as the unusual event was determined to be low risk.
Figure 4: Timeline of DIGEST incident scoring, where the score remained low as the unusual event was determined to be low risk.

These examples show the value of enhanced scoring. DIGEST helps teams act sooner on the threats that count and spend less time chasing the ones that do not.

The next phase of advanced detection is here

Darktrace understands what incidents look like. We have seen, investigated, and learned from them at scale, including over 90 million investigations in 2024. With DIGEST, we can share our deep understanding of incidents and their behaviors with you and triage these incidents using Cyber AI Analyst.

Our ability to innovate in this space is grounded in the maturity of our team and the experiences we have built upon in over a decade of building AI solutions for cybersecurity. This experience, along with our depth of understanding of our data, techniques, and strategic layering of AI/ML components has shaped every one of our steps forward.

With DIGEST, we are entering a new phase, with another line of defense that helps teams prioritize and reason over incidents and threats far earlier in an incident’s lifecycle. DIGEST understands your incidents when they start, making it easier for your team to act quickly and confidently.

DIGEST is available in Darktrace 6.3, along with a new embedding model – DEMIST-2 – designed to provide reliable, high-accuracy detections for critical security use cases.

[related-resource]

Continue reading
About the author
Margaret Cunningham, PhD
Director, Security & AI Strategy, Field CISO
Your data. Our AI.
Elevate your network security with Darktrace AI