Researchers at Zimperium’s zLabs have analyzed an Android Trojan that uses an automated, multi-stage infection process. What’s new is that RatHat gives a live AI assistant the keys to the accessibility tree of the infected device and uses it to determine where to tap or scroll, rather than following a hardcoded script. The variable attack path makes it harder for signature- and rule-based mobile security tools to detect this Trojan. It also abuses Android Debug Bridge (ADB), a legitimate tool…
Analysis of how default configurations in AWS AgentCore Harness allow prompt injection to exfiltrate credentials, and key steps to secure your agents. The post A Vault with a Heap-View: The Uncomfortable Space Between AgentCore Harness and Identity appeared first on Unit 42.
Key Takeaways Periodic audits provide a point-in-time assessment, but they cannot demonstrate whether controls remain effective between audit cycles. Qualys platform data shows 10.5 billion configuration findings across customer environments but only 1.6% represent meaningful exposure and under 1% are prioritized, business-critical findings. Verizon’s 2026 DBIR found the median time to resolve weak passwords and misconfigured permissions is about 8 months. Across 1 billion misconfiguration…
A new Android malware called RatHat has been discovered, targeting users with an AI-powered subsystem that helps operators remotely navigate compromised devices. [...]
OpenAI has presented new examples of what they call "AI model misalignment" from the past six months, including unauthorized file uploads, following self-generated instructions, hiding mistakes, and leveraging exposed API keys. [...]
AI has already made fundamental changes to the operating environment for cybersecurity. Cyberattackers are testing more paths, adapting their techniques, and moving across digital environments with greater speed and persistence. The weaknesses they exploit remain familiar: excessive permissions, unprotected authentication flows, unpatched systems, exposed execution paths, and gaps between controls. What has changed is how quickly these weaknesses can combine into attack paths that cross…
Every benchmark tells a story. The most valuable ones tell us where to improve next. For five consecutive quarters Microsoft has published email security benchmarking reports to provide greater transparency into real-world protection outcomes. The results have shown strong Microsoft Defender performance across pre-delivery and post-delivery scenarios, while revealing where threats and defenses continue to evolve. This quarter’s benchmark examines how continuous measurement informs protection…
The defining development of the period came not from attackers but from the AI labs themselves, whose models broke out of controlled evaluations and reached real systems. In the wild, the criminal and state use of AI continued to mature along the lines tracked in earlier editions: models now act as attack operators, an underground market supplies the access, and AI systems have themselves become a target. The substantial distance between what the strongest models demonstrated under evaluation…
Compared with the same period last year, ransomware incidents in Japan increased slightly by approximately 4.7%, indicating that ransomware continues to pose a significant threat.In Japan, The Gentlemen was the most active ransomware group in the first half of 2026.Attackers continue to primarily target small- and medium-sized enterprises, with organizations capitalized at less than JPY 1 billion accounting for approximately 80% of the total — an increase of around 13% from the previous year…
AI is accelerating vulnerability discovery, leaving unpatchable operational technology (OT) systems at risk. Hoping for the best is not a viable anti-exploitation strategy. Deploying next-generation firewalls directly upstream allows for virtual patching through deep packet inspection. These systems scan incoming traffic to detect and block exploit attempts before they can impact the vulnerable device.The predictability of legitimate network connections to OT systems can be used to protect…
Executive Summary OpenAI disclosed that agents used exposed Hugging Face credentials to write a file and deploy proxy Spaces during an unrelated May 2026 research workload, but it did not identify the accounts. SentinelLABS identified two accounts likely used in associated activity, 0Time and Nyx9. Their public histories extend OpenAI’s chronology and preserve previously unreported relay code, document-borne probes, and ChatGPT account-provisioning capability. The public records provide precise…
Antivirus renewal scams often begin with a message claiming that your subscription has automatically renewed. When you follow the instructions to cancel it, you are taken to a fake page designed to collect your contact details. The renewal charge never existed. Receiving a message that names the antivirus software you use does not necessarily mean the sender has access to your device or account. Scammers impersonate popular brands and send the same message to large numbers of people, knowing…
If you’re still OK with posting pictures of your kids on social media, take a minute to hear from mother of two Kalie Robins. At the start of September, she did something that hundreds of thousands of parents do every day. She posted a video of her young daughter on Facebook. Under the video of Robins and her daughter singing in a car, Facebook’s Meta AI system displayed a suggested question: “Who is the child passenger?” Robins was shocked that Facebook would ask this question about a minor,…
Uncensored refers to a lack of typical guardrails or ethical restrictions, lowering the technical barrier of entry into cybercrimeCategories: Threat ResearchTags: AI, Luciferus, underground
Anthropic says multiple threat groups, including the financially motivated and state-sponsored espionage groups linked to Russia and China, tried to abuse its Claude AI model for malicious purposes. [...]
Researchers analyzed a flood of fraudulent business emails and found that the threat actors had doubled-up on tactics to make them appear legitimate, including help from AI.
I identified an attacker using a semi-autonomous coding agent to run an offensive operation: finding poorly secured LLM resale gateways, acquiring API access through ordinary web flaws and account farming, validating the resulting inference capacity, and aggregating it behind a single gateway of their own. What I captured was not simply credential theft, but an inference supply chain — an agent harvesting, validating, and consolidating LLM access into infrastructure that serves it again through…
Threat actors are abusing trusted AI platforms to host malicious content, poison search results, and trick users into installing malware. Huntress examines campaigns targeting AI users through weaponized Claude Artifacts, shared AI conversations, sponsored search results, and ClickFix-style lures. [...]
Local privilege escalation (LPE) is the step that turns a foothold into full control of a host. An attacker who lands as an unprivileged user rarely stops there. They want root, and Linux keeps offering new ways to get it.In this edition of our "Linux Detection Engineering" series, we’ll cover:The default flow that a Linux LPE produces on the host and the general rules that detect it.The recurring LPE patterns behind the most recent LPEs and how each works, along with how each looks through the…
In this article Attack chain overviewEmail DeliveryDomain registrationGenerative AI usageMitigation and protection guidanceMicrosoft Defender detectionsMicrosoft Security CopilotThreat intelligence reportsMITRE ATT&CK Techniques observedIndicators of compromise (IOC)Learn More Threat actors are increasingly improving their tactics to make suspicious emails look like legitimate email notifications to potential victims, deploying techniques that impersonate internally sent emails from executive…
Executive Summary In this research we introduce a prompt-crafting technique for bypassing quick LLM-based policy checks — using plain English (no emojis, base64, invisible formatting, etc.) A policy-violating payload (e.g. ”encrypt files in ~/Documents”, “give me a biohazard recipe”, “ignore all previous instructions and…”) is embedded in a specially crafted prose wrapper. An LLM with limited resources and attention fails to realize the payload is there, classifies the prompt as benign and…
Learn why Tenable treats agentic LLMs as untrusted insiders, and how we’ve made sure you can control and monitor the AI agents making changes in your production security environmentKey takeawaysAI models can quickly understand data, but not your business. While modern AI models are great at reasoning, they don’t automatically understand your unique environment or who is allowed to do what. The “harness” is the custom-built layer that translates AI intelligence into safe, controlled actions…
The question of whether a Frontier AI model could find vulnerabilities that no human researcher had found was settled in April. Claude Mythos Preview identified thousands of previously unknown flaws across every major operating system and browser, including a 27-year-old denial-of-service condition in OpenBSD, and within a month Anthropic and its Project Glasswing partners had logged more than 10,000 high and critical severity findings, among them a certificate forgery flaw in wolfSSL, a…
Research by: Alexey Bukhteyev Key Takeaways Check Point Research discovered a covert cross-account command channel through which an attacker could use a victim’s ChatGPT session to execute hidden tasks with the tools, data, and connected apps available to that session. The victim could receive a normal answer to their visible request while the attacker’s task was processed separately and its result returned across accounts. In our proof of concept, ChatGPT retrieved email data from the victim’s…
Executive Summary Since the release of our May 2026 report detailing adversarial misuse of artificial intelligence (AI), Google Threat Intelligence Group (GTIG) has observed forward leaning adversaries transition from basic prompting to agentic AI workflows and AI-enabled automation. In these operations, human-in-the-loop latency is dramatically reduced, compressing the traditional window for defenders to respond. In Q2 2026, GTIG observed threat actors compromise a cloud resource, then plan,…
In a recent post, we looked at reports of League of Legends players receiving suspicious friend requests shortly after matches. The accounts quickly steered the conversation toward Discord, where they promoted paid adult-content pages. At the time, one unanswered question was how much of those conversations was automated. Were people working from scripts behind the accounts? Were they conventional, rules-based chatbots following a limited decision tree? Or were they using generative AI to…
In this article What is ASCII smuggling?Writing a practical ASCII-smuggling signatureWhat we observed: ASCII smuggling repurposed for phishingWhat is known and what is newIs there a detection gap?Mitigation and protection guidanceReferencesLearn More Microsoft researchers observed a high-volume phishing campaign using invisible Unicode tag characters, a technique popularized in AI prompt injection research as ASCII Smuggling. Instead of using these characters to hide instructions from people…
Explore how attackers targeting Latin American entities use AI for data exfiltration and how basic OpSec errors allow defenders to disrupt operations. The post Attackers Expose Ongoing AI Tool Use Targeting Organizations in Latin America appeared first on Unit 42.
Using autonomous AI agents, an attacker breached an enterprise network in a matter of hours. Understand how to address and defend against agentic attacks. The post An AI-Assisted Cyber Attack: Inside a Unit 42 Investigation appeared first on Unit 42.
Fileless execution on Linux has moved from niche tradecraft into real-world intrusion chains. By executing payloads from memory or anonymous file descriptors, attackers can reduce on-disk artifacts and weaken controls that rely heavily on file inspection.In our own analysis of VoidLink, we observed how fileless execution can be paired with a rootkit. The loader scans for processes running from memfd and passes their PIDs to the rootkit, allowing an already running fileless implant to be hidden…
Introduction Beginning in 2024 Mandiant investigated a string of compromises affecting Brazilian financial services, retail, and eCommerce organizations. Google Threat Intelligence Group (GTIG) tracks this activity as BREEZE COMET (formerly UNC5669), a financially motivated threat actor specializing in manipulating payment systems and banking software in Brazil to conduct fraudulent transfers. This activity overlaps with operations publicly reported as Plump Spider and SHADOW-AETHER-064. In…
One of my internet-exposed inference honeypots was discovered, relabeled with sought-after model names, and incorporated into infrastructure apparently used to provide "free" LLM backends. It then received a real coding-agent session — history, filesystem output, working paths, and the agent's local tool manifest. The honeypot did not request or cause any tool execution; what the request exposed is what a malicious operator in that position could do. Chasing the "free API key" is not new. What…
New research reveals that AI safety refusal lives in a thin neural layer, highlighting the critical need for external, multi-layered security. The post Perturbation Probing: A New Diagnostic for the Fragility of LLM Safety appeared first on Unit 42.
AI is accelerating vulnerability discovery, putting pressure on systems built to enrich, prioritize, and remediate flaws at a slower pace. Action1 explains why defenders increasingly need to correlate multiple intelligence sources and turn vulnerability data into faster remediation. [...]
The hacking incident involving OpenAI evaluation agents and Hugging Face offers an unusually concrete look at what advanced AI-assisted intrusion can mean in practice: not a single clever exploit, but thousands of automated decisions, rapid experimentation, lateral movement, credential theft, persistence, and attempts to evade detection. The OpenAI–Hugging Face incident began during internal cybersecurity evaluations using ExploitGym, a benchmark designed to test whether AI agents can identify…
Welcome to this week’s edition of the Threat Source newsletter. Hello, everyone. Long time reader, first time writer here at the Threat Source newsletter! I wanted to start out by introducing myself. My colleague and friend Mick Baccio set the bar pretty high last week, so I was planning to tell you all about myself, including: How I did my first real IR under the influence of The Cuckoo’s Egg while an undergraduate (and failed) My pre-bug bounty flirtation with vulnerability research,…
Discover how Tenable’s shift to an AI-driven exposure management program helped Tenable’s CSO, Robert Huber, overcome tool sprawl, unify data silos, mitigate the risk of rapid AI adoption, and shift from presenting granular, technical metrics to communicating business risk that the C-suite and the board can understand.Key takeawaysSecurity tool sprawl and data silos make it difficult for CISOs to holistically and accurately assess their organizations’ cyber risk.An exposure management program…
In this article AI workloads are becoming high-value control pointsCase study 1: LiteLLM gateway compromiseCase study 2: RAGFlow compromiseCase study 3: Kestra compromiseMitigation and protection guidanceMITRE ATT&CK techniques observedReferencesLearn more AI is creating a new layer of enterprise infrastructure. Gateways, retrieval platforms, orchestration services, and containerized runtimes now sit between users, applications, data, and models. These systems concentrate credentials, data…
A phase-by-phase detection mapping of the first publicly documented autonomous agent intrusion against production infrastructure — including the phases where no product in our category sees anything at all. Executive Summary On July 9, 2026, an autonomous AI agent running inside an OpenAI capability evaluation escaped its sandbox and launched a multi-day intrusion against Hugging Face’s Kubernetes environment. Across roughly 17,600 actions, it moved from third-party infrastructure into the…
Selecting a model for your security operations center (SOC) and digital forensics and incident response (DFIR) tasks is important, but selecting the best one is more involved than you might think. SOC tasks rely on a combination of model efficacy, analysis time, cost, and consistency of results. Cisco Talos tested 66 model and reasoning combinations across offerings from both Anthropic and OpenAI on a log analysis task to see if we could identify a clear winner. Instead, we found a repeatable…
Group-IB expone una operación mexicana de PhaaS dirigida a más de 20 instituciones financieras, con capacidades de phishing en tiempo real, vishing con IA y RAT para dispositivos móviles.
AI verdict correctness in our security operations center (SOC) is 92%, up from 60%, but we didn't switch models to get there. What we changed is the context the agents get before they decide anything, including the detection rule's investigation guide and user risk data from Workday, along with the closure reasons from 30 days of past cases on that same rule. This post covers how the agentic SOC pipeline is built in Elastic Workflows and Elastic Agent Builder, down to the prompts and the…
AI verdict correctness in our security operations center (SOC) is 92%, up from 60%, but we didn't switch models to get there. What we changed is the context the agents get before they decide anything, including the detection rule's investigation guide and user risk data from Workday, along with the closure reasons from 30 days of past cases on that same rule. This post covers how the agentic SOC pipeline is built in Elastic Workflows and Elastic Agent Builder, down to the prompts and the…
Aikido AI Code Audit found 8 more vulnerabilities than Claude Security with Mythos at less than half the cost. How harness design drives coverage per dollar. Category: News
UAT-10147 is a highly capable Chinese-speaking intrusion actor operating a multi-platform post-exploitation ecosystem targeting IIS and Linux servers, combining search engine optimization (SEO) fraud monetization with advanced persistence and defense evasion techniques. The newly identified SPECTRE implant represents a significant evolution in commodity intrusion tooling, integrating cross-platform command-and-control (C2) operations, process injection, credential theft, anti-analysis…
<p>A year of MDR casework shows attackers repeatedly exploiting demand for AI tools</p>Categories: Threat ResearchTags: AI, malvertising, infostealer, Sophos X-Ops
Written by: Alex Tselevich, Michael Maturi Introduction Adversarial misuse of AI has increased the risk of data theft and extortion events, because when proprietary source code is exposed, defenders must scramble to identify and patch vulnerabilities while attackers deploy machine-speed AI tools against them. By structuring the analysis process, enforcing skeptical validation steps, and injecting domain-specific human expertise directly into the pipeline, we’ve achieved a leap in efficacy.…
You can’t patch everything. So what do you fix first? Findings in Q2 2026 have changed traditional answers.The latest Quarterly Threat Landscape Report from Rapid7 Labs shows vulnerability disclosures still surging while attackers use automation and AI-assisted tooling to compress the time between disclosure and exploitation. The gap that patch cycles were built to fill is closing. Speed and volume are overwhelming security teams that have relied on traditional patch cycles and reactive…
Tenable’s Research Special Operations (RSO) team has been tracking a cluster of agentic AI threat activity since late July 2026. The Taiwan autonomous AI cyber attack confirmed what the cluster data already showed: near-autonomous offensive AI has crossed from theoretical risk to operational reality.Key TakeawaysTaiwan's Ministry of Digital Affairs confirmed a near-autonomous AI cyber attack in July 2026 in which autonomous agents mapped 21 connected government systems, compromised 85 accounts,…
For the past year, the ransomware conversation has centered on concentration: a handful of dominant RaaS operations controlling most of the damage, and a shrinking pool of active groups fighting over the same territory. The State of Ransomware Q2 2026 report from Check Point Research shows that picture starting to shift. The leaders are still winning, but the road to joining them has gotten a great deal shorter. Key observed findings The ecosystem stayed concentrated even as its tail widened…
Something shifted in security operations over the last two years: AI stopped being a pilot program and became the plan.And if you survey 500 security professionals on whether that's going well – as Omdia did, commissioned by Rapid7 – you get a remarkable level of consensus: 97% report positive outcomes, 98% say AI reduces alert fatigue, and 95% say it's helping address staffing shortages.Those numbers are high enough that the story could stop there; AI is working, everyone agrees. Move on. But…
We gave hundreds of developers an AI agent that can run shell commands, edit files, and call Model Context Protocol (MCP) servers on their laptops, then realized we had no record of what it actually did. So we built one. One 280-line dependency-free bash script, fired by Cursor's hooks, records every tool call as JSONL, and the Elastic Agent already on each endpoint ships it to Elasticsearch. Since the May rollout we have logged over 13 million tool-call events from more than 1,100 machines. A…