SPS Digital Tech @SPSDigitalTech

EvilAI Malware Posing as Legit AI Tools A malware campaign called EvilAI is masquerading as legitimate AI or productivity software. These trojans present polished interfaces, valid certificates, and plausible functionality to trick users. Once installed, they sit quietly, steal…

AIOpsDoom: Manipulating AI-Driven IT Operations via Telemetry Attacks A recent academic study revealed AIOpsDoom, an attack that targets AI-driven IT operations (AIOps). By injecting manipulated telemetry (logs, metrics, sensor data), attackers trick LLM-based agents into…

PromptLock: The First AI-Powered Malware Prototype Security firm ESET discovered PromptLock, a novel ransomware that uses the gpt-oss:20b model to generate malicious Lua scripts on the fly. It doesn’t require downloading the full AI model; instead, it tunnels commands via a…

State-Sponsored Exploitation of Libraesva Email Gateway A command injection vulnerability was discovered in the Libraesva Email Security Gateway, a trusted perimeter tool for filtering malicious mail. Attackers exploited the flaw by sending crafted compressed attachments,…

AI Agents Vulnerable to Hijacking Attacks Researchers have demonstrated that AI agents (automated systems designed to perform tasks like retrieving data, moving files, or executing workflows) are highly vulnerable to hijacking. Unlike one-time prompt injections, these attacks…

“Lies-in-the-Loop” Attack Defeats AI Coding Agents A new attack method called Lies-in-the-Loop shows how AI coding assistants can be manipulated into writing insecure code. Instead of one malicious input, attackers gradually feed misleading instructions into the AI’s workflow,…

Okta and Vercel v0: AI-Generated Phishing Sites in 30 Seconds Hackers exploited Vercel’s v0 generative AI tool, which is designed to build websites from natural language prompts. Instead of legitimate use, attackers used it to generate phishing sites mimicking Okta’s login…

ShadowLeak: Zero-Click Exploit in ChatGPT Deep Research Agent A vulnerability called ShadowLeak was uncovered in ChatGPT’s Deep Research agent. Attackers could send a specially crafted email that triggered the AI to process and leak data in the background without any user…

HexStrike-AI Framework and Rapid Exploitation of Citrix Vulnerabilities HexStrike-AI is an open-source AI-powered framework originally intended for red teaming and security testing. It integrates many AI agents and orchestrates over 150 cybersecurity tools via its “orchestration…

Cursor IDE: “CurXecute” & “MCPoison” Remote Code Execution Vulnerabilities Cursor, an AI-assisted code editor used widely by developers, had two serious vulnerabilities: CurXecute and MCPoison. These flaws allowed attackers to execute code remotely. In particular, a malicious…

Multi-Agent Systems Hijacked to Execute Malicious Code In a recent academic study, multi-agent AI systems (where agents coordinate to carry out workflow tasks powered by LLMs) were shown to be vulnerable even when individual agents resisted malicious prompts. Attackers could…

Lenovo Chatbot Vulnerability: Prompt Injection in Customer Service Workflow Lenovo’s customer service AI chatbot “Lena” had a security flaw: using a crafted prompt (~400 characters) that combined HTML, JSON and plain text instructions, an attacker could inject malicious code…

AIOps Workflow Compromise via Telemetry Manipulation Researchers uncovered that AI-for-IT operations (AIOps) systems, which automate incident detection and remediation, can be subverted via manipulated telemetry data. An adversary sends crafted, error-inducing signals or…

AI-generated code in “vibe coding” workflows containing security flaws A study by Veracode looked at over 100 large language models (LLMs) across ~80 coding tasks. They found that about 45% of AI-generated code contains security flaws, even if the code seems production-ready.…

Millions of Bouygues Telecom Users’ Data Exposed French telecom Bouygues Telecom confirmed that data of 6.4 million of its 27 million customers was compromised. Exposed details included contact info, civil status, contract details, and for business customers, IBANs, but not…

Jaguar Land Rover’s Cyber Woes Might Last Until October JLR suffered a cyberattack that is disrupting production and sales, with plants in UK and supplier facilities affected. They now admit that “some data” was impacted.

LNER Passenger Data Breach via Third-Party Supplier (UK) London North Eastern Railway (LNER) disclosed that a third-party supplier’s systems were hacked, exposing customer contact details and journey history. No financial or passwords compromised.

Cyberattack Hits Luxury Resort Just Before Major Tournament A ransomware group called SafePay breached the K Club resort in Co Kildare, Ireland, just before the Irish Open. Some financial, IT, and administrative files were stolen and later leaked.

23andMe Seeks Approval for $50M Breach Settlement After a 2023 breach exposed genetic & personal data of ~6.4 million users, 23andMe is asking a judge to approve a $50 million settlement (up from a prior $30 million) to resolve over 250,000 claims. The settlement includes a…

Mac Users Targeted via Fake Microsoft Teams aka Odyssey Stealer Cybercriminals are luring macOS users with fake Microsoft Teams website links that deliver the new “Odyssey” stealer malware.