> ## Content Index
> Fetch the complete content index at: https://www.cybrsecmedia.com/llms.txt
> Use this file to discover other available public pages before exploring further.

# Distributed Malware: The Next Challenge for EDR
- URL: https://www.cybrsecmedia.com/the-next-malware-threat-wont-hide-from-your-security-tools-it-will-make-them-lose-the-plot/
- Published: 2026-11-18T13:48:29.000Z
- Updated: 2026-11-18T13:51:37.000Z
- Description: Texas A&M researcher Marcus Botacin warns that distributed malware and covert communication channels could overwhelm behavioral detection by making malicious activity visible but nearly impossible to connect.
- Author: Bill Brenner
- Tags: Distributed Malware Detection, EDR evasion, behavioral malware detection, Article

*This article is based on Marcus Botacin's discussion with Michael Farnum and Phillip Wylie on CYBR.HAK.CAST, building on his CYBR.SEC.CON 2026 research presentation.* ***Full episode and related article:***

[Free to Dream with Marcus BotacinIn this episode of CYBR.HAK.CAST, hosts Michael Farnum and Phillip Wylie welcome Texas A&M professor and cybersecurity researcher Marcus Botacin to explore the future of malware detection, distributed attacks, and the evolving challenges facing security professionals. Building on Marcus's CYBR.SEC.CON. 2026 presentation, they examine![](https://storage.ghost.io/c/ab/67/ab676516-71e3-473d-8f73-9e0692f5aaee/content/images/icon/CYBR.SEC.Media-Logo-copy-70c83d48-824a-4ef8-a714-a7e405d9c532.jpg)CYBR.SEC.Media, and CYBR.SEC.Media![](https://storage.ghost.io/c/ab/67/ab676516-71e3-473d-8f73-9e0692f5aaee/content/images/thumbnail/Marcus-Botacin_Ghost-3b9df61e-cc63-41cd-b3d0-f514ad8cd047.png)](https://www.cybrsecmedia.com/free-to-dream-with-marcus-botacin/)

[The Future of Malware Defense: Beyond DetectionTexas A&M researcher Marcus Botacin argues that the future of malware defense may depend less on identifying every malicious action and more on disrupting attacks before they succeed. His research points toward hardware-based antivirus and a fundamental rethink of endpoint security.![](https://storage.ghost.io/c/ab/67/ab676516-71e3-473d-8f73-9e0692f5aaee/content/images/icon/CYBR.SEC.Media-Logo-copy-ccfd4816-2156-4a13-9314-06997d3cc333.jpg)CYBR.SEC.MediaBill Brenner![](https://storage.ghost.io/c/ab/67/ab676516-71e3-473d-8f73-9e0692f5aaee/content/images/thumbnail/c81654b9-4f20-4985-90b2-9d2dd86beadf-a971d139-5f75-4e4f-a9b2-690a35b28211.png)](https://www.cybrsecmedia.com/what-if-detecting-malware-is-the-wrong-fight-a-researcher-wants-to-change-the-rules/)

Security teams have spent years improving their ability to see what happens inside an endpoint. More telemetry. Better behavioral analysis. More sophisticated detection rules. The assumption is that if defenders can see enough malicious activity, they can eventually connect the dots and stop an attack.

But what happens when attackers deliberately scatter those dots so widely that connecting them becomes practically impossible?

That's the problem Texas A&M University professor and cybersecurity researcher Marcus Botacin is investigating. During a recent episode of CYBR.HAK.CAST with hosts Michael Farnum and Phillip Wylie, Botacin described how distributed malware could challenge one of the fundamental assumptions behind modern endpoint security: that malicious behavior can be reconstructed by following the actions of an individual process or executable.

The emerging threat isn't necessarily malware that hides from security tools. It's malware that makes its activity visible while concealing the relationships that make those activities malicious.

That could have serious implications for endpoint detection and response (EDR), security operations centers and the next generation of malware defense.

[Subscribe to the CYBR.SEC.Media newsletter](#/portal/signup/free)

## When the attack is bigger than any single process

Botacin explored the concept during his presentation at CYBR.SEC.CON 2026, titled *You Can See Me, But You Can't Track Me: Evading Behavioral Detection with Distributed Malware and Covert Synchronization Channels.*

Traditional behavioral detection works by observing a sequence of activities. A malicious executable launches, accesses resources, modifies files, communicates with other systems or performs other actions that collectively reveal its purpose.

Security products attempt to recognize those sequences through behavioral rules, machine learning and other detection techniques.

But Botacin asks what happens when no single executable performs enough of the attack to reveal the broader objective.

Imagine an attacker delivering three separate malicious components through three different channels. One arrives through a phishing email. Another through a browser download. A third through a collaboration platform such as Slack.

Each component performs only part of the attack.

Individually, their activities may appear unremarkable. Together, they accomplish something malicious.

The challenge for defenders is no longer simply identifying suspicious behavior. It's reconstructing the relationships among processes that may have no obvious connection.

As Botacin explained during the podcast, the question becomes whether defenders can correlate events scattered across multiple processes and surrounded by unrelated system activity.

The individual events are visible. The attack emerges only when those events are understood together.

## A denial-of-service attack against detection itself

One of the more provocative aspects of Botacin's research is his description of distributed malware as something resembling a denial-of-service attack against the detector.

Not because the malware necessarily floods the security product with network traffic or crashes its processes.

Instead, it creates a correlation problem.

If malicious activities are distributed across numerous processes, defenders must determine which events belong together, in what order they occurred and whether their combined behavior represents an attack.

That task becomes increasingly difficult as the number of processes and unrelated system events grows.

In conventional behavioral detection, an analyst or automated system might identify a sequence of actions — A, B, C and D — as a recognizable malicious pattern.

In Botacin's distributed model, those actions could occur in separate processes, potentially on separate CPU cores, surrounded by thousands of unrelated events.

The defender must reconstruct the sequence without necessarily knowing which activities are connected.

And the attacker can make that task even harder by changing how the individual components communicate.

## Three ways malware could disappear into the noise

Botacin's research examines progressively more evasive approaches to coordinating distributed malicious processes.

The first involves conventional synchronization mechanisms, including application programming interfaces, mutexes and semaphores.

These mechanisms allow separate processes to coordinate their activities. They also provide defenders with potential opportunities to observe the relationships.

Existing EDR tools and logging systems could evolve to monitor these interactions and reconstruct the attack sequence.

The second approach involves shared-memory synchronization.

Rather than relying on conventional API calls, malicious components coordinate through memory-based mechanisms. That removes some of the observable activity security products traditionally use to understand relationships between processes.

Detecting these interactions could require more comprehensive memory monitoring, introducing significant performance and architectural challenges.

The third approach is potentially the most disruptive: covert communication through hardware side channels.

In Botacin's research, malicious processes can coordinate through mechanisms involving CPU cache behavior rather than explicit software communication.

That creates a fundamentally different detection challenge.

If the processes aren't exchanging recognizable messages or using observable synchronization APIs, conventional monitoring may have little information with which to reconstruct their relationships.

"The disruption point is not simply, oh, I split the malware in multiple process, but the disruption is I break the synchronization between them," Botacin explained.

In other words, the attacker isn't just fragmenting the malicious behavior. The attacker is concealing the evidence that the fragments belong together.

## Why more telemetry may not solve the problem

For security teams, the immediate instinct might be to collect more data.

If an EDR platform cannot reconstruct the attack, expand its visibility. Capture more events. Monitor additional system interactions. Apply more sophisticated analytics.

Botacin acknowledges that this approach could work against less sophisticated forms of distributed malware.

When malicious components coordinate through conventional APIs, for example, defenders can potentially expand monitoring to identify the communication.

But covert synchronization introduces a different problem.

If the relationship between malicious processes is not exposed through conventional software activity, collecting additional logs may not reveal it.

That doesn't mean behavioral detection becomes obsolete. It means some forms of attack could challenge the assumptions on which existing detection methods operate.

Botacin's work also raises difficult questions about the computational cost of reconstructing increasingly complex attack sequences.

Security products must distinguish malicious coordination from the enormous volume of legitimate activity occurring across modern operating systems.

The more possible relationships they must evaluate, the harder that task becomes.

And attackers don't necessarily need to make every component invisible. They need to make the overall operation difficult enough to recognize that the attack can proceed.

## The stealthier the malware, the more fragile the attack

There is an important limitation to this emerging threat, and Botacin is careful to explain it.

More sophisticated synchronization techniques introduce tradeoffs for attackers.

Conventional API-based coordination is relatively straightforward and reliable, but easier for defenders to observe.

Shared-memory coordination can be harder to detect, but introduces additional operational requirements.

CPU cache-based side channels may offer greater stealth, but they also depend on specific system conditions.

If those conditions change, the attack may fail.

As Botacin explained, increasing stealth can reduce the reliability of the malicious operation. A cache flush, process interruption or other change in system behavior could disrupt the coordination needed for the attack to succeed.

That creates opportunities for defenders, even when the underlying malicious activity cannot be fully reconstructed.

It also complicates the idea that more evasive malware is automatically more dangerous.

An attack that is extremely difficult to detect but equally difficult to execute reliably presents a different risk than one that is both stealthy and resilient.

The research suggests attackers may need to balance those competing requirements as they develop more sophisticated distributed techniques.

## A warning about where malware could go next

Botacin isn't claiming that covertly synchronized distributed malware has already become a widespread threat.

His research is about anticipating how attackers might evolve beyond techniques that existing security products are designed to recognize.

He expects any movement toward more sophisticated distributed malware to occur gradually.

Attackers could begin with conventional synchronization mechanisms, prompting defenders to improve correlation and visibility. As those defenses mature, attackers could explore less observable coordination methods.

That progression would create another chapter in the long-running competition between offensive techniques and defensive technologies.

For security leaders, the immediate lesson isn't to abandon behavioral detection or rush to replace existing EDR platforms.

It's to question how well those platforms recognize malicious activity distributed across multiple processes — and how dependent their detection capabilities are on observing explicit relationships between those processes.

Botacin's research also offers a useful direction for red and purple teams: evaluate whether existing controls can recognize coordinated malicious behavior when no individual component reveals the complete attack.

The larger issue is whether security architectures designed around individual processes and recognizable behavioral sequences can adapt to attacks deliberately constructed to defeat those assumptions.

For years, the security industry has operated on the premise that greater visibility leads to better detection.

Botacin's research suggests the next generation of malware may test the limits of that belief.

The security tools might see everything the attacker does. The challenge will be understanding what it all means.

[![](https://storage.ghost.io/c/ab/67/ab676516-71e3-473d-8f73-9e0692f5aaee/content/images/2026/11/2027-Events-Banner-35.png)](https://www.cybrsecmedia.com/conference/)