VMware Servers Hacked Five Days After Patch

VMware vCenter Server is centralised management software that IT teams use to control entire fleets of virtual machines and servers from one place. According to Bleeping Computer, a critical flaw in its Syslog component is being actively exploited in the wild.

The vulnerability, CVE-2026-59310, is a directory traversal flaw (a weakness that lets an attacker navigate to files on a server they are not permitted to access) in vCenter's Syslog server. An attacker with basic network access, no login required, can trigger it to run any code they choose.

Broadcom disclosed and patched the flaw on July 29. By August 3, attackers were already in. By August 5, over 340 compromised servers had connected to attacker-controlled infrastructure. The final count reached 361 victim IP addresses across 47 countries.

Once inside, the attackers deployed reverse_ssh, an open-source tool that creates an outbound connection back to the attacker. Because the connection goes out rather than in, it slips past firewalls watching for incoming threats.

Incident response firm QUIRSO believes an advanced persistent threat (APT) group is behind the campaign. Broadcom has released no workarounds. A patch is the only fix.

What to do: If your organisation runs VMware vCenter, update immediately to version 9.1.0.0300, 9.0.2.0100, 8.0 U3k, or 8.0 U2f depending on your branch.

Anthropic recently switched on invisible watermarks in everything Claude writes, weaving the mark into the model's word choices rather than hiding it in file metadata. Within days, a wave of "watermark remover" tools appeared online, according to Bleeping Computer.

The tools can genuinely strip hidden characters and file metadata. The actual watermark, however, lives in which words the model chose, and no public tool can reliably remove that today. The only known method is a full rewrite by a second AI model, which defeats the point of using the original. Anthropic has not yet released a public detector, so none of the claims can be independently verified.

Some commercial tools advertise "clean, undetectable output" while testing their results against generic AI detectors rather than Anthropic's actual watermark system.

What to do: If you are evaluating AI content for authenticity, treat "watermark removed" claims with scepticism until Anthropic publishes its detection tool.

What VMware vCenter Server is: Centralised IT management software that lets organisations control, monitor, and configure all their virtual machines and servers from a single platform.

What it is: A directory traversal flaw (a weakness that lets an attacker request files outside the intended folder structure) in vCenter's Syslog server component, leading to remote code execution (RCE — running attacker-supplied commands on the target machine).

Who's at risk: Any organisation with a vCenter server reachable over the network. No login is required to trigger the flaw.

CVSS: 9.8 (Critical — patch today).

Root cause: The Syslog server component fails to validate file path inputs, allowing a crafted request to escape the intended directory and reach sensitive system areas. There are no workarounds. Only patching resolves it.

Attack vector: An attacker with network access sends a specially crafted request to the vCenter Syslog server. The server processes the malformed path without restriction, granting file access and enabling arbitrary code execution. In active exploitation, attackers then drop the reverse_ssh tool, which phones home to attacker infrastructure, bypassing inbound firewall rules.

Detection strategies:

1. Look for unexpected outbound SSH connections from vCenter hosts to unknown external IPs.
2. Search for reverse_ssh binaries using QUIRSO's published YARA rule — but validate any hits, as the tool also has legitimate uses.
3. Check for new, unrecognised processes running under the vCenter service account.

Recommended actions:

1. Apply the emergency patch immediately: 9.1.0.0300, 9.0.2.0100, 8.0 U3k, or 8.0 U2f.
2. Restrict network access to vCenter — it should never be publicly internet-facing.
3. Review outbound connection logs for traffic to unfamiliar infrastructure since August 3.

WhatsApp is rolling out a new feature called Scam Alert, according to Bleeping Computer. It runs a machine learning model directly on your device to flag messages from unknown contacts that match known scam patterns.

Nothing leaves your phone for analysis. No message content is sent to WhatsApp, Meta, or anyone else. If the model spots something suspicious, you will see a prompt to block, report, or continue. If it gets it wrong, you can mark the chat as trusted.

The feature is in limited beta right now but is expected to roll out more broadly. To check if it is available: open WhatsApp, go to Settings → Privacy and look for Scam Alert. Enable it if you see it.

Scam messages via messaging apps are one of the most common ways people are defrauded. This is a free, privacy-respecting layer of protection that takes thirty seconds to enable.

Seventeen draft technical standards for the EU's Cyber Resilience Act (a law requiring connected products sold in Europe to meet mandatory cybersecurity requirements) are now open for review, according to Help Net Security. The standards cover products including smart home assistants, password managers, and connected toys. Manufacturers who follow an approved standard gain a presumption of legal compliance. The comment window closes between mid-September and mid-November 2026, with the full obligation kicking in at the end of 2027. Small businesses selling any connected hardware or software into European markets should read the relevant draft now, not after the deadline.