Cybersecurity Brief – 2026-05-17

Sun, 17 May 2026 00:00:00 +0000

Major Incidents or Breaches

Grafana disclosed that an unauthorized party obtained a token granting access to its GitHub environment, resulting in the download of its codebase and an extortion attempt against the company ^[1].
Instructure’s Canvas platform was impacted by a ransomware incident, which has now concluded ^[6].
OpenAI workers were affected by a supply chain attack, though specifics regarding the method or scope were not detailed ^[6].

A critical vulnerability in the Funnel Builder plugin for WordPress is under active exploitation, enabling attackers to inject malicious JavaScript into WooCommerce checkouts for skimming payment information ^[2].
Proof-of-concept code has been published for a critical vulnerability in NGINX Plus and NGINX open source, which was introduced in 2008 and only recently patched ^[7].
A security researcher alleged that Microsoft quietly fixed a critical Azure Backup for AKS vulnerability without issuing a CVE, after initially rejecting the report. Microsoft disputes these claims ^[3].

The Russian threat group Secret Blizzard has evolved the Kazuar backdoor into a modular peer-to-peer botnet, enhancing its capabilities for long-term persistence, stealth operations, and data collection ^[4].
An alleged darknet market kingpin was arrested, though further operational details were not provided ^[6].

AI agents are increasingly being used to discover and exploit obscure vulnerabilities, while the surge in AI-generated code is contributing to a growing attack surface and new security challenges for defenders ^[5].

Sat, 02 Aug 2025 00:00:00 +0000

Akira ransomware has been observed exploiting SonicWall SSL VPN devices in a surge of attacks since late July 2025, reportedly targeting fully-patched devices and suggesting exploitation of a likely zero-day vulnerability.
Pi-hole disclosed a data breach resulting from exploitation of a vulnerability in the GiveWP WordPress donation plugin, exposing donor names and email addresses.
A malicious npm package, @kodane/patch-manager, generated using AI, was identified and removed after it was used to drain Solana cryptocurrency funds from over 1,500 victims.
Microsoft is investigating whether the ToolShell exploit was leaked via the Microsoft Active Protections Program (MAPP).
Russian state-sponsored APT group Secret Blizzard (Nobelium) conducted adversary-in-the-middle (AitM) attacks at the ISP level targeting foreign embassies in Moscow, leading to malware infections on diplomatic devices.
An Apple ID phishing scam in Ohio led to an in-person theft of $27,000 from a victim.

A high-severity remote code execution vulnerability, dubbed CurXecute, affecting almost all versions of the AI-powered Cursor code editor, has been patched. The flaw allowed attackers to execute commands via prompt injection.
SonicWall SSL VPN devices are reportedly being exploited via a previously unknown (zero-day) vulnerability in ongoing Akira ransomware attacks.
Recent vulnerabilities in Microsoft SharePoint Server have been exploited by threat actors to deploy bespoke command-and-control frameworks.

Akira ransomware operators are actively exploiting SonicWall SSL VPN devices, with evidence suggesting the use of a zero-day vulnerability.
Threat actors are using fake Microsoft OAuth applications, leveraging the Tycoon Kit, to breach Microsoft 365 accounts and facilitate credential harvesting.
The threat actor Storm-2603 is exploiting Microsoft SharePoint vulnerabilities to deploy a DNS-controlled backdoor using a custom C2 framework (AK47 C2) as part of Warlock and LockBit ransomware campaigns.
Russian APT Secret Blizzard is conducting sophisticated AitM attacks against foreign diplomatic targets in Moscow.

Increased use of AI by attackers, including the creation of malicious npm packages for cryptocurrency theft.
Prompt injection remains a practical and exploitable attack vector against AI-powered developer tools.
Only about half of the code generated by large language models (LLMs) is considered secure, raising ongoing concerns about the security of AI-generated code.
The SIEM market is undergoing transformation as XDR platforms and generative AI impact security analytics.
New malware techniques such as “Shade BIOS” allow malware to persist and operate independently of the operating system, evading traditional detection and remediation.
Cybercriminals are increasingly targeting Gen Z, recognising young, digitally native workers as a distinct and vulnerable attack surface.

The US Department of Homeland Security (DHS) has launched over $100 million in funding to strengthen community cyber defences.
A bipartisan US Senate bill has been introduced to create a national strategy for quantum cybersecurity migration, aimed at preparing federal agencies for quantum computing threats.
ISC2 has launched a new security certificate programme focused on AI expertise, covering AI fundamentals, ethics, and associated risks.
Microsoft has increased the maximum reward for its .NET bug bounty programme to $40,000 for valid reports of remote code execution or privilege escalation vulnerabilities.
Meta is sponsoring the upcoming Pwn2Own Ireland 2025 hacking contest, with a $1 million reward for a demonstrated zero-click WhatsApp exploit.