The images in this article were generated with artificial intelligence. How we publish
Botnets don't disappear; they reinvent. In recent weeks, the response and detection teams have refocused on a malicious software piece known as Chaos, but not because it is something completely new: the remarkable thing is that it has changed the stage. This malware, which had already been detected a while ago affecting routers and edge devices, has started to take advantage of cloud environments that are poorly configured - a jump that increases its reach and its damage potential.
To understand why this matters it is necessary to review what Chaos is and how it has evolved. When it appeared in the reports of some cybersecurity firms, it was described as a multiplatform malware capable of running remote orders, downloading additional modules, replicating by gross force against SSH keys, mining cryptomonedas and launching service denial attacks (DDoS) by multiple protocols. This combination made it particularly versatile: it could live in both Windows and Linux systems and use different techniques to take advantage of committed resources.

The novelty now is that attackers are directing that flexibility to cloud services that present configuration errors. Recent investigations point to intrusions in deliberately vulnerable instances - such as an exposed Hadoop that allowed remote execution - where attackers manufacture an "application" that actually runs a sequence of commands to download a malicious binary, give it extensive permissions (e.g. a classic chmod 777), run it and then delete the file to make it difficult to forensic analysis.
This type of manoeuvre is not new in itself, but it is the objective: the transit from poorly configured edge devices and containers to cloud deployments exposes organizations that rely on managed platforms but have not properly hardened permits, access controls and supervision. Signatures like Darktrace have documented detections in their lure networks, and intelligence groups like Lumen Black Lotus Labs They were among the first to describe Chaos' original capabilities.
An interesting technical detail of the new binary is that, although it retains many of the functionalities that researchers already knew, it eliminates certain old routines intended to spread by SSH or to exploit vulnerabilities in routers. Instead it incorporates a SOCKS proxy capacity that transforms the committed team into a relay for malicious traffic. That changes the rules of the game: the infected system ceases to be just a mining machine or to participate in DDoS and becomes an anonimization infrastructure for the rest of the criminal activity, which complicates the attribution and containment of operations that originate from the botnet.
The movement to offer proxy services fits a broader trend in the cybercrime market: monetization of bots is no longer limited to cryptomoneda mining or rental for DDoS attacks, but includes the sale of access, tunnels and anonymity services that other threats can take advantage of. This type of development has also been observed in other botnet families, which have been adding modules to diversify their income and their usefulness to third parties.
On attribution, as always, the certainties are scarce. Some indirect findings - such as the presence of Chinese characters in code or the use of infrastructure located in China - suggest that the authors could operate from that region, but that is only a clue. The infrastructure reused by the malicious actors can be mixed with phishing campaigns or other operations: in one of the reported incidents, the domain from which the Chaos agent was downloaded had already been linked to a mail campaign that delivered spy programs and decoy documents, a work that commercial researchers then labelled as Operation Silk Lure.
What can a security team do to this kind of threat? The first thing is to accept that poorly configured clouds are an attractive target and that monitoring must be adapted to that risk. It is essential to reduce the exposed surface: restrict management interfaces, review permit policies, use strong authentication in SSH and API accesses, and control the creation of applications or jobs that allow arbitrary commands to run. At the same time, detection should include less obvious signals, such as unusual binary discharge patterns, cascade-changing commands, or outgoing connections that establish proxy tunnels. Traffic analysis tools, web application firewalls and cloud detection solutions help to raise early alerts.

The security community regularly publishes guides and notices on how to mitigate botnets and risks associated with the cloud; organisms such as CISA and research projects offer resources to be followed, and threat intelligence firms often update Commitment Indicators and practical recommendations. Consult these reports and implement access patches, rules and policies as soon as possible significantly reduces the likelihood that a legitimate deployment will end up being part of a botnet.
In short, the case of Chaos is a reminder that criminal tools evolve and explore new attack surfaces. An uncontrolled cloud architecture is an attractive input vector, "and the incorporation of proxies into botnets makes committed teams facilitators of other malicious actions. Defense requires not only technology, but processes and operational discipline: good hygiene in configurations, continuous monitoring and rapid response when foreign behaviors appear.
If you want to deepen the background and technical analysis, it is appropriate to review the reports and blogs of companies investigating these threats - for example, publications of Darktrace, the intelligence analysis of Lumen Black Lotus Labs and reports of specialized laboratories such as Seqrit Labs- and to complement this reading with media and technical coverage in specialized media such as Bleeping Computer to follow the evolution of the case.
Related
More news on the same subject.

FBI and six countries link Integrity Technology Group to entity post theft in SE Asia
On October 8, the FBI and agencies in six countries issued a joint warning that assigns to a Chinese company, Integrity Technology Group, a sustained series of intrusions whose ...

Campaign with LLM and ARTEX attacks South Korean financial institutions and exfilters data
Security researchers have documented a campaign directed against South Korean financial institutions using language-driven attack tools to automate intrusions and data extractio...

ChainDrop campaign exposes tensorlake in npm; version 0.5.144 withdrawal
A package of npm called tensorlake, an SDK in TypeScript oriented to Tensorlake applications and services, was engaged in a supply chain campaign linked to the attack family kno...

Google reports DNS kidnapping: TLS certificates for google.com.gh, google.sl and google.as
Google reported on October 6 that attackers managed to issue unauthorized HTTPS certificates for Google and YouTube names after compromising authoritative DNS records of three t...

Cyber risk in 2026 moves to workflows and IA, according to Voice of the CISO
The data added by five editions of the Voice of the CISO study - including the most recent findings of 2026 - draw a less intense change than risk location: the threat is moving...

Phishing BitB points to advertising professionals and account managers to steal MFA
Security researchers have described a phishing campaign for advertising professionals and account managers that uses a human-operated platform to mimic ad products linked to IA ...