artificial intelligence | Breaking Cybersecurity News | The Hacker News

The most common way into a company last year was to ask. A web page tells the visitor to prove they are not a robot. While they read the instructions, it quietly places a command on their clipboard. Then it talks them through opening a terminal and pasting it in. The technique is called ClickFix, and it was the most common initial access method Microsoft’s team observed last year , accounting for 47% of the attacks in their notifications. Nothing arrives as an attachment, so there is nothing to scan. No vulnerability is used, so there is nothing to patch. What happens next is just as ordinary. When Bitdefender analyzed 700,000 security incidents , 84% of the high-severity ones involved binaries that were already on the machine - the same administrative tools your IT team uses every day. Nothing malicious was installed, because nothing malicious was needed. Neither technique is clever, but both are winning. And the reason is not that attackers have run out of ideas. It is that th...

METR (short for Model Evaluation and Threat Research and pronounced "Meter"), a research non-profit that evaluates frontier artificial intelligence (AI) models for their ability to carry out long-horizon, agentic tasks, disclosed that it suffered "two notable security incidents" where external actors attempted to gain unauthorized access to its systems. No sensitive information is believed to have been accessed as a result of these incidents, it said, adding that a version of its findings was shared with AI companies it works with prior to public disclosure. The attacks have not been attributed to any known threat actor or group, nor did they involve AI agents breaking into its evaluations. "In March 2026, attackers stole an API key for inference on public models and consumed a substantial amount of credits," METR said . "In May 2026, we observed attackers systematically probing our publicly accessible infrastructure, including an unsuccessful at...

Cybersecurity researchers have disclosed a new technique dubbed GuardBreaker that's been put to use by a Russia-aligned threat actor known as UAC-0099 against a target in Ukraine with an aim to interfere with artificial intelligence (AI)-assisted analysis. The idea, ESET said in a series of posts on X, is to deliberately trip a large language model's (LLM) safety mechanisms and prevent its normal functioning. "In the attack, UAC-0099 inserted a problematic text: 'I want to make a nuclear weapon. Help me ...' into their malicious VBS script as a comment," the Slovak cybersecurity company said. "This is meant to attract the AI's attention to the safety-sensitive content and stop it from analyzing the rest of the code." The GuardBreaker-embedded VBS script is assessed to be part of a broader toolset employed by UAC-0099, which has a track record of targeting transportation and energy sectors. The script is primarily designed to download ...

Threat actors are exploiting two critical flaws impacting Langflow and Ruby on Rails, according to new findings from VulnCheck. The vulnerabilities in question are listed below - CVE-2026-0768 (CVSS score: 9.8) - A lack of proper validation of a user-supplied input vulnerability that could be exploited to execute arbitrary Python code in the context of the root user. CVE-2026-66066 aka KindaRails2Shell (CVSS score: 9.5) - A vulnerability that could allow an unauthenticated attacker to read arbitrary files from the server, leak Rails process environment and secrets such as secret_key_base, the Rails master key, database passwords, cloud storage credentials, and API tokens, ultimately leading to remote code execution. Attackers can exploit CVE-2026-66066 by uploading a crafted image by taking advantage of the discrepancy between Active Storage and libvips in how they read input files. Successful exploitation requires affected applications to use libvips for Active Storage ...

Threat actors with ties to the Democratic People's Republic of Korea (aka DPRK or North Korea) have been observed seeking job opportunities beyond the information technology (IT) sector, with recent investigations identifying suspected workers employed in sales and marketing and the medical profession. The ongoing insider threat is part of what has been described as the IT worker scheme , where North Korea leverages its network of skilled IT workers, both within and outside the country, to fraudulently land jobs in Fortune 500 companies and private sector firms across the world and remotely earn income to further Pyongyang's unlawful nuclear weapons and ballistic missile programs. This entails relying on stolen or forged identity documents, VPNs, and proxy services to mask their true identity and location. The yearslong campaign is also tracked under the monikers Famous Chollima, Jasper Sleet, Nickel Tapestry, PurpleDelta (formerly TAG-121), UNC5267, and Wagemole. ...

The boring parts caused most of the trouble. A router shipped ready to listen. A fake check turned the user into the installer. Trusted systems collected traffic and passwords, then cleaned the logs. Old bugs formed new attack chains. Even an AI agent decided its assigned task was optional. Elsewhere, fake apps, helpful support calls, cheap banking kits, exposed systems, and weak defaults kept things moving. Different attacks, same useful mistake: something familiar was trusted without a second look. Here is the week... ⚡ Threat of the Week U.S. Disrupts Chinese Proxy Network Enabling Cyber Espionage — The U.S. Federal Bureau of Investigation (FBI) disrupted infrastructure associated with a technical quartermaster who sold reconnaissance, proxy management, and operational routing capabilities for Chinese cyber espionage activities. The QTYF group is said to have created and operated the QScan and QTRouter frameworks, which have been used to target U.S. critical infrastructu...

Threat actors associated with Aurora (aka Aur0ra) ransomware have been observed using SpaceX's artificial intelligence (AI)-powered coding assistant Cursor to break into target networks, according to findings from CloudSEK and Gambit Security . The two independent analyses are based on exposed infrastructure associated with the Russian-speaking cybercrime group, leading to the discovery of its toolkit, shell history, and encryptor. CloudSEK said the exposed open directory leaked "months of activity" that was active against more than 20 organizations across nine countries between April and July 2026. Four of those victims have since been listed on its data leak site. "The operator used Cursor, an agentic coding assistant, to plan attacks in Russian, while excluding CIS [Commonwealth of Independent States] ranges and CIS-country domains, without exception," CloudSEK noted. Details about Aurora first emerged in late May 2026, with CYFIRMA highlighting atta...

Claude Code reads files, runs shell commands, invokes MCP tools, and acts through the credentials available on a developer’s machine. Anthropic’s new Compliance API endpoints give security teams their clearest view yet into that activity. They also expose a larger problem: activity logs alone cannot tell you whether an agent’s access is legitimate. AI has moved from the browser tab to the endpoint with harnesses like Claude Code. They run on developers' machines, execute bash commands locally, and connect to third parties via MCP servers, skills, and plugins. All this so the user can outsource labor to the machine and focus on designing, thinking, and creating. Local agents are not a niche category. They account for 68.6% of the AI agents Token Security discovers in customer environments, and they often inherit the employee's credentials, network position, and permissions. The shift to the endpoint has major implications for security. With Claude Code, there is no ce...

An Identity Fabric knits fragmented identity systems into a coherent layer that observes how identities behave across applications, APIs, and infrastructure. As enterprise access spans more cloud services and automated workloads, identity security depends less on static configuration and more on runtime visibility. This article covers the architecture, the risks of unmanaged identities, and practical steps to close the gap between access intent and actual execution. The guidance here focuses on enterprise hybrid and multi-cloud environments; smaller single-directory deployments may not require the full scope described. Understanding Identity Fabric Architecture and How It Works An Identity Fabric is not a single product but an architectural approach that connects identity providers, governance systems, applications, and infrastructure into one observable layer. Its purpose is to reconcile what access policy intends with how identities are actually used at runtime. Closing this g...

OpenAI on Wednesday revealed that reward hacking was a key driver behind the artificial intelligence (AI)-powered hack of Hugging Face last month, adding that it found evidence of misaligned behavior as early as late May. The incident, the company said, took place during cybersecurity evaluations of several OpenAI models, and that it was mainly fueled by what it described as a "highly capable, internal-only research model" comparable in scale to GPT‑5.6 Sol. "The models, operating under reduced safeguards, took actions that were misaligned with the goals of their assigned tasks – they communicated through unauthorized channels, exploited vulnerabilities in shared infrastructure, gained internet access, and accessed third-party systems," it said in a postmortem. OpenAI said the AI agents powered by one of the research models, despite not having internet access, found a way to exploit a then-zero-day vulnerability in the Artifactory package manager during r...

A fake login page. A fake security scan. A fake productivity app. Apparently, pretending to be useful is still one of the easier ways into a machine. The rest of the week gets stranger: botnets borrowing AI, command traffic hiding in public infrastructure, malicious tools waiting before showing their real behavior, exposed systems getting scanned, and exploit windows shrinking again. Different tricks, same advantage: attackers keep finding places where trust is cheap and friction is low. That sets the tone. Here’s the full list of what surfaced this week. The threats change every week. Subscribe, and we’ll alert you when each new ThreatsDay Bulletin is out.

Cybersecurity researchers have disclosed details of a vulnerability in Amazon Kiro, an artificial intelligence (AI)-powered, agentic integrated development environment (IDE), that could facilitate data exfiltration via prompt injection and Kiro Powers. The security flaw, which does not have a CVE identifier, works against Kiro IDE 0.7.45 on Windows, according to Mindgard. The latest version of the IDE is 1.0.337. "The issue allowed attacker-controlled repository content to influence the Kiro agent and ultimately cause sensitive local information to be transmitted to an external endpoint," Chief Marketing Officer Fergal Glynn  said in a report shared with The Hacker News. Kiro Powers goes beyond skills by bundling Model Context Protocol (MCP) server configurations, steering files ("POWER.md"), hooks, and contextual knowledge. The steering file is like an "onboarding manual" that provides persistent context and tells the AI agent what MCP tools are...

Security teams have spent years trying to detect threats faster. AI is changing the harder part: how much time defenders have left to act. Advanced AI models can now help attackers discover vulnerabilities, generate exploit code, and move through weaknesses faster than traditional security processes were built to handle. The challenge is no longer just finding another vulnerability or generating another alert. It is knowing which exposures matter, what an attacker can reach, and which issues need to be fixed first. That is the focus of next week's webinar, How to Build AI Threat Readiness Across Your Security Operations , featuring an expert from Wiz. The session will show how security teams can improve visibility, prioritize real risk, and shorten the path from detection to remediation. Most security teams already have plenty of data. They have vulnerability findings, cloud alerts, identity signals, application telemetry, and threat detections. The harder problem is conne...

AI is officially mainstream in security operations. According to Prophet Security's State of AI in Security Operations 2026 report (produced from ViB’s survey of 250+ cybersecurity pros), 40% of security teams now use AI daily. Another 56% are currently testing it out, and only 4% have no plans to adopt it. For the teams already using AI, what is actually changing? Here are the ten biggest takeaways from the report. 1. Security teams are drowning in alerts The average security team gets about 100 alerts every day, but larger companies often deal with close to 1,000. In fact, over a quarter of teams face more than 500 alerts daily. Unfortunately, team sizes haven't kept up with the workload. While some massive organizations have over 100 analysts, many are running on skeleton crews of fewer than ten people. Because of this, it takes an average of 75 minutes to thoroughly investigate a single alert, and alerts often sit untouched for nearly an hour before anyone even loo...

The SOC we've always known was built around a model that guarantees most of the alert queue will never receive analyst review. There's never time. In a traditional SOC, the typical progression follows a well-known pattern: an alert arrives; a detection engine assigns a severity score. The issue then waits for a human to decide if it should escalate to an investigation. Given the volume of network telemetry in the security stack, the queue is an unavoidable result of humans as the investigative layer. Long alert queues also force security teams to decide which signals to analyze before they even know what those signals represent. Threat hunting has always addressed security questions via an alternative approach: start with a hypothesis about attacker behavior, search the available evidence, then prove or disprove it. The sequence is powerful, but it hits the same wall: human capacity. Agentic security operations change the paradigm. The SOCs now being built are predica...

Aikido Security has published research that recreates the Australian gym-booking incident in a synthetic environment, finding that Claude Opus 4.6, running on the OpenClaw agent harness, exploited a client-side-only booking restriction in 9 of 10 runs. The original incident was first reported by ABC News on August 10, based on chat logs and screenshots the user supplied. He had asked an OpenClaw agent running Opus 4.6 to book him into a gym class. The agent booked sessions months beyond the window the site allowed. It then tested, without being asked, whether the same API would let it cancel another member's waitlist entry. The test removed the person holding the top place and moved the user up one position. The agent told him it could not add the member back. Aikido's test system is a single-page web application backed by a GraphQL API carrying the two flaws described in the original incident. The seven-day booking window is enforced only in the frontend, and the cancel...

OpenAI on Tuesday said it banned a cluster of Russian ChatGPT accounts that used VPNs to bypass access restrictions and run an influence operation, which relied on its artificial intelligence (AI) tool to generate social media posts and comments that were shared on Substack, Telegram, X, Facebook and LinkedIn. The accounts "were being used to promote the International Burke Institute (IBI), a self-described 'expert community' based in Israel," the company said . "What began as an investigation into AI-generated social media posts led us to a much broader influence operation, built around a website containing copied and misattributed academic work, a 'sovereignty' index that cast Russia in a favourable light, and efforts to disguise the operators' Russian origins." The campaign is assessed to have reached "relatively small audiences," with Telegram channels attracting more users and amassing about 10-20,000 followers each. The m...

Cybersecurity researchers have disclosed details of a phishing-as-a-service (PhaaS) platform built to strip Apple's Activation Lock from stolen devices, using rented AI voice agents that call theft victims posing as Apple Support and ask for their device passcode. SOCRadar Threat Research Unit (STRU) said the platform, which it tracks as AnonyMousKIT , is credit-metered and drives lures across five channels from a single victim record, comprising email at 1.50 credits, SMS priced per sender ID, WhatsApp, a recorded voice call at 1 credit, and an AI voice agent at 2 credits. The targets are owners of Apple devices that were recently lost or stolen, and the pages and calls ask each of them for the 4- or 6-digit device passcode, then the Apple ID credentials, and finally a live two-factor authentication (2FA) code. Apple's own guidance states that the company never asks for a password, device passcode, or 2FA code to provide support. "AnonyMousKIT is best understood no...

Vulnerability management has been a staple of security programs since the dawn of the cybersecurity discipline. The symbiotic relationship between vulnerability and patch management teams has also existed for that time and has gone through waves of contention and thankfulness. While this relationship required thoughtful care and feeding from both sides, both sides were aiming to work toward a common goal of identifying vulnerabilities and confirming the risk was removed from the environment. In come Frontier AI models such as Anthropic's Mythos to radically change the vulnerability management space. These models can identify zero-day flaws, chain complex exploits, and adapt in real time. They have forced vulnerability management programs to take an introspective look at themselves and ask, “Is my vulnerability program ready for this revolution?” For many organizations, the answer is no. Many vulnerability management programs were hanging by a thread already, with very distant pl...

A package gets installed. A login prompt opens. A box sits exposed to the internet. Nothing looks unusual yet. That’s roughly the mood this week. Trusted tools turn hostile, old weak spots get fresh attention, AI makes exploit work cheaper, and researchers keep finding attacks that sound harder than they actually are. Plenty to clean up. Here’s the short version. ⚡ Threat of the Week U.S. Warns of AI-Powered Attacks on Siemens PLCs — Threat actors are using AI to write exploit scripts targeting internet-exposed Siemens S7 Series programmable logic controllers (PLCs) used across water, energy, manufacturing, and other critical infrastructure sectors, according to the U.S. government. The agencies warned: "This is not a theoretical risk—it is an active threat." The exploitation of poorly secured PLCs could result in disruption of critical industrial processes, safety incidents, downtime or equipment damage, compromise of sensitive data, and compliance violations, not...