artificial intelligence | Breaking Cybersecurity News | The Hacker News

Few technologies have moved from experimentation to boardroom mandate as quickly as AI. Across industries, leadership teams have embraced its broader potential, and boards, investors, and executives are already pushing organizations to adopt it across operational and security functions. Pentera’s AI Security and Exposure Report 2026 reflects that momentum: every CISO surveyed reported that AI is already in use across their organizations. Security testing is inevitably part of that shift. Modern environments are too dynamic, and attack techniques too variable, for purely static testing logic to remain sufficient on its own. Adaptive payload generation, contextual interpretation of controls, and real-time execution adjustments are necessary to get closer to how attackers, and increasingly their own AI agents, operate. For experienced security teams, the need to incorporate AI into testing is no longer in question. You have to fight fire with fire. Wh...

OpenAI on Tuesday unveiled GPT-5.4-Cyber , a variant of its latest flagship model, GPT‑5.4 , that's specifically optimized for defensive cybersecurity use cases, days after rival Anthropic unveiled its own frontier model, Mythos . "The progressive use of AI accelerates defenders – those responsible for keeping systems, data, and users safe – enabling them to find and fix problems faster in the digital infrastructure everyone relies on," OpenAI said . In conjunction with the announcement, the artificial intelligence (AI) company said it's ramping up its Trusted Access for Cyber ( TAC ) program to thousands of authenticated individual defenders and hundreds of teams responsible for securing critical software. AI systems are inherently dual-use, as bad actors can repurpose technologies developed for legitimate applications to their own advantage and achieve malicious goals. One core area of concern is that adversaries could invert the models fine...

Cybersecurity researchers have unmasked a novel ad fraud scheme that has been found to leverage search engine poisoning (SEO) techniques and artificial intelligence (AI)-generated content to push deceptive news stories into Google's Discover feed and trick users into enabling persistent browser notifications that lead to scareware and financial scams. The campaign, which has been found to target the personalized content feeds of Android and Chrome users, has been codenamed Pushpaganda by HUMAN's Satori Threat Intelligence and Research Team. "This operation, named for push notifications central to the scheme, generates invalid organic traffic from real mobile devices by tricking users into subscribing to enabling notifications that presented alarming messages," researchers Louisa Abel, Vikas Parthasarathy, João Santos, and Adam Sell said in a report shared with The Hacker News. At its peak, about 240 million bid requests have been associated wit...

OX Security recently analyzed 216 million security findings across 250 organizations over a 90-day period. The primary takeaway: while raw alert volume grew by 52% year-over-year, prioritized critical risk grew by nearly 400%. The surge in AI-assisted development is creating a "velocity gap" where the density of high-impact vulnerabilities is scaling faster than remediation workflows. The ratio of critical findings to raw alerts nearly tripled, moving from 0.035% to 0.092%. Key Findings from the 2026 Analysis: CVSS vs. Business Context: Technical severity scores are no longer the primary driver of risk. The most common elevation factors were High Business Priority (27.76%) and PII Processing (22.08%) . In modern environments, where a vulnerability lives is now more important than what the vulnerability is. The AI Fingerprint: We observed a direct correlation between the adoption of AI coding tools and the quadrupling of critical f...

Monday is back, and the weekend’s backlog of chaos is officially hitting the fan. We are tracking a critical zero-day that has been quietly living in your PDFs for months, plus some aggressive state-sponsored meddling in infrastructure that is finally coming to light. It is one of those mornings where the gap between a quiet shift and a full-blown incident response is basically non-existent. The variety this week is particularly nasty. We have AI models being turned into autonomous exploit engines, North Korean groups playing the long game with social engineering, and fileless malware hitting enterprise workflows. There is also a major botnet takedown and new research proving that even fiber optic cables can be used to eavesdrop on your private conversations. Skim this before your next meeting. Let’s get into it. ⚡ Threat of the Week Adobe Acrobat Reader 0-Day Under Attack   — Adobe released emergency updates to fix a critical...

Anthropic restricted its Mythos Preview model last week after it autonomously found and exploited zero-day vulnerabilities in every major operating system and browser. Palo Alto Networks' Wendi Whitmore warned that similar capabilities are weeks or months from proliferation. CrowdStrike's 2026 Global Threat Report puts average eCrime breakout time at 29 minutes. Mandiant's M-Trends 2026 shows adversary hand-off times have collapsed to 22 seconds.  Offense is getting faster. The question is where exactly defenders are slow — because it's not where most SOC dashboards suggest. Detection tooling has gotten materially better. EDR, cloud security, email security, identity, and SIEM platforms ship with built-in detection logic that pushes MTTD close to zero for known techniques. That's real progress, and it's the result of years of investment in detection engineering across the industry.  But when adversaries are operating on timelines measured in s...

While much of the discussion on AI security centers around protecting ‘shadow’ AI and GenAI consumption, there's a wide-open window nobody's guarding: AI browser extensions.  A  new report from LayerX exposes just how deep this blind spot goes, and why AI extensions may be the most dangerous AI threat surface in your network that isn't on anyone's radar. AI browser extensions don't trigger your DLP and don't show up in your SaaS logs. They live inside the browser itself, with direct access to everything your employees see, type, and stay logged into. AI extensions are 60% more likely to have a vulnerability than extensions on average, are 3 times more likely to have access to cookies, 2.5 times more likely to be able to execute remote scripts in the browser, and 6 times more likely to have increased their permissions in the past year. These extensions install in seconds and can remain...

Thursday. Another week, another batch of things that probably should've been caught sooner but weren't. This one's got some range — old vulnerabilities getting new life, a few "why was that even possible" moments, attackers leaning on platforms and tools you'd normally trust without thinking twice. Quiet escalations more than loud zero-days, but the kind that matter more in practice anyway. Mix of malware, infrastructure exposure, AI-adjacent weirdness, and some supply chain stuff that's... not great. Let's get into it. Resilient hybrid botnet surge Phorpiex Botnet Detailed A new variant of the botnet known as Phorpiex (aka Trik) has been observed, using a hybrid communication model that combines traditional C2 HTTP polling with a peer-to-peer (P2P) protocol over both TCP and UDP to ensure operational continuity in the face of server takedowns. The malware acts as a conduit for encrypted payloads, ma...

As AI tools become more accessible, employees are adopting them without formal approval from IT and security teams. While these tools may boost productivity, automate tasks, or fill gaps in existing workflows, they also operate outside the visibility of security teams, bypassing controls and creating new blind spots in what is known as shadow AI. While similar to the phenomenon of shadow IT, shadow AI goes beyond unapproved software by involving systems that process, generate, and potentially retain sensitive data. The result is a category of risk that most organizations are not yet equipped to govern: uncontrolled data exposure, expanded attack surfaces, and weakened identity security. Why shadow AI is spreading so quickly Shadow AI is expanding rapidly across organizations because it is easy to adopt and instantly useful, yet largely unregulated. Unlike traditional enterprise software, most AI tools require little to no setup, allowing employ...

The Fragmented State of Modern Enterprise Identity Enterprise IAM is approaching a breaking point. As organizations scale, identity becomes increasingly fragmented across thousands of applications, decentralized teams, machine identities, and autonomous systems.  The result is Identity Dark Matter: identity activity that sits outside the visibility of centralized IAM and beyond the reach of security teams. According to Orchid Security ’s analysis , 46% of enterprise identity activity occurs outside centralized IAM visibility. In other words, nearly half of the enterprise identity surface may be operating unseen. This hidden layer includes unmanaged applications, local accounts, opaque authentication flows, and over-permissioned non-human identities. It is further amplified by disconnected tools, siloed ownership, and the rapid rise of Agentic AI. The consequence is a widening gap between what the security organizations think th...

Artificial Intelligence (AI) company Anthropic announced a new cybersecurity initiative called Project Glasswing  that will use a preview version of its new frontier model, Claude Mythos , to find and address security vulnerabilities. The model will be used by a small set of organizations, including Amazon Web Services, Apple, Broadcom, Cisco, CrowdStrike, Google, JPMorgan Chase, the Linux Foundation, Microsoft, NVIDIA, and Palo Alto Networks, along with Anthropic, to secure critical software. The company said it's forming this initiative in response to capabilities observed in its general-purpose frontier model that demonstrate a "level of coding capability where they can surpass all but the most skilled humans at finding and exploiting software vulnerabilities." Because of its cybersecurity capabilities and concerns that they could be abused, Anthropic has opted not to make the model generall...

In the rapid evolution of the 2026 threat landscape, a frustrating paradox has emerged for CISOs and security leaders: Identity programs are maturing, yet the risk is actually increasing . According to new research from the Ponemon Institute , hundreds of applications within the typical enterprise remain disconnected from centralized identity systems. These "dark matter" applications operate outside the reach of standard governance, creating a massive, unmanaged attack surface that is now being aggressively exploited—not just by human threat actors, but by autonomous AI agents . The Invisible Threat: Disconnected Apps & AI Amplification Modern enterprises have invested heavily in IAM and Zero Trust, but the "last mile" of identity—legacy apps, localized accounts, and siloed SaaS—remains a stubborn blind spot. The entry of AI into the workforce has turned this gap from a compliance headache into a critical vul...

A high-severity security vulnerability has been disclosed in Docker Engine that could permit an attacker to bypass authorization plugins ( AuthZ ) under specific circumstances. The vulnerability, tracked as CVE-2026-34040 (CVSS score: 8.8), stems from an incomplete fix for CVE-2024-41110 , a maximum-severity vulnerability in the same component that came to light in July 2024. "Using a specially-crafted API request, an attacker could make the Docker daemon forward the request to an authorization plugin without the body," Docker Engine maintainers said in an advisory released late last month. "The authorization plugin may allow a request which it would have otherwise denied if the body had been forwarded to it." "Anyone who depends on authorization plugins that introspect the request body to make access control decisions is potentially impacted." Multiple security vulnerabilities, including Asim Viladi Oglu Manizada, Cody, Oleh Konko, and Vladimir...

New academic research has identified multiple RowHammer attacks against high-performance graphics processing units (GPUs) that could be exploited to escalate privileges and, in some cases, even take full control of a host. The efforts have been codenamed GPUBreach , GDDRHammer , and GeForge . GPUBreach goes a step further than GPUHammer , demonstrating for the first time that RowHammer bit-flips in GPU memory can induce much more than data corruption and enable privilege escalation, and lead to a full system compromise. "By corrupting GPU page tables via GDDR6 bit-flips, an unprivileged process can gain arbitrary GPU memory read/write, and then chain that into full CPU privilege escalation — spawning a root shell — by exploiting memory-safety bugs in the NVIDIA driver," Gururaj Saileshwar, one of the authors of the study and Assistant Professor at the University of Toronto, said in a post on LinkedIn. What makes GPUBreach notable is that it works eve...

Threat actors are exploiting a maximum-severity security flaw in Flowise , an open-source artificial intelligence (AI) platform, according to new findings from VulnCheck. The vulnerability in question is CVE-2025-59528 (CVSS score: 10.0), a code injection vulnerability that could result in remote code execution. "The CustomMCP node allows users to input configuration settings for connecting to an external MCP (Model Context Protocol) server," Flowise said in an advisory released in September 2025. "This node parses the user-provided mcpServerConfig string to build the MCP server configuration. However, during this process, it executes JavaScript code without any security validation." Flowise noted that successful exploitation of the vulnerability can allow access to dangerous modules such as child_process (command execution) and fs (file system), as it runs with full Node.js runtime privileges. Put differently, a threat actor who weaponizes the flaw can execu...

This week had real hits. The key software got tampered with. Active bugs showed up in the tools people use every day. Some attacks didn’t even need much effort because the path was already there. One weak spot now spreads wider than before. What starts small can reach a lot of systems fast. New bugs, faster use, less time to react. That’s this week. Read through it. ⚡ Threat of the Week Axios npm Package Compromised by N. Korean Hackers —Threat actors with ties to North Korea seized control of the npm account belonging to the lead maintainer of Axios, a popular npm package with nearly 100 million weekly downloads, to push malicious versions containing a cross-platform malware dubbed WAVESHAPER.V2. The activity has been attributed to a financially motivated threat actor known as UNC1069. The incident demonstrates how quickly the compromise of a popular npm package can have ripple effects through the ecosystem. T...

The latest ThreatsDay Bulletin is basically a cheat sheet for everything breaking on the internet right now. No corporate fluff or boring lectures here, just a quick and honest look at the messy reality of keeping systems safe this week. Things are moving fast. The list includes researchers chaining small bugs together to create massive backdoors, old software flaws coming back to haunt us, and some very clever new tricks that let attackers bypass security logs entirely without leaving a trace. We are also seeing sketchier traffic on the underground and the usual supply chain mess, where one bad piece of code threatens thousands of apps. It is definitely worth a quick scan before you log off for the day, if only to make sure none of this is sitting in your own network. Let's get into it. Pre-auth RCE chain exposed Security Flaws in Progress ShareFile watchTower Labs has disclosed two securi...

In December 2025 , we shared the first-ever The State of Trusted Open Source report, featuring insights from our product data and customer base on open source consumption across our catalog of container image projects, versions, images, language libraries, and builds. These insights shed light on what teams pull, deploy, and maintain day to day, alongside the vulnerabilities and remediation realities these projects face. Fast forward a few months, and software development is accelerating at a pace that most didn’t see coming. AI is increasingly embedded across the development lifecycle, from code generation to infrastructure automation, as models become more advanced and better at meeting the demands of modern work. This shift is expanding what teams can build and how quickly they can ship. It is also reshaping the security landscape. Before diving into the numbers, it’s important to explain how we perform this analysis. We examined over 2,20...

The Computer Emergency Response Team of Ukraine (CERT-UA) has disclosed details of a new phishing campaign in which the cybersecurity agency itself was impersonated to distribute a remote administration tool known as AGEWHEEZE. As part of the attacks, the threat actors, tracked as UAC-0255 , sent emails on March 26 and 27, 2026, posing as CERT-UA to distribute a password-protected ZIP archive hosted on Files.fm and urged recipients to install the "specialized software." The targets of the campaign included state organizations, medical centers, security companies, educational institutions, financial institutions, and software development companies. Some of the emails were sent from the email address "incidents@cert-ua[.]tech." The ZIP file ("CERT_UA_protection_tool.zip") is designed to download malware packaged as security software from the agency. The malware, per CERT-UA, is a remote access trojan codenamed AGEWHEEZE.  A Go-based malware, AGEWHEEZE...

There is a character that keeps appearing in enterprise security departments, and most CISOs know exactly who that is. It doesn’t build. It doesn’t enable. Its entire function is to say "No." No to ChatGPT. No to DeepSeek. No to the file-sharing tool the product team swears by. For years, this looked like security. But in 2026, "Doctor No" is no longer just a management headache – it is a systemic security liability. Because when you block the work, users don’t stop. They reroute. The Tax-Evaders of Productivity When security feels like a tax on efficiency, employees find a way to "evade" it. The industry has long relied on Endpoint Agents to enforce control. But as any CISO knows, these agents come with a heavy "tax." They hook into the OS kernel, they’re invasive, they notoriously break during macOS updates, and they make high-performance machines run hot. The result? Users find workarounds. Files move into personal Gmail. Prompts are...