artificial intelligence | Breaking Cybersecurity News | The Hacker News

Google on Thursday announced that it fixed a whopping 1,072 security bugs in Chrome versions 149 and 150, surpassing the total number of flaws the company fixed across the prior 23 milestones combined. Both versions were released last month. In its latest patch for Chrome 151, released Wednesday, the tech giant resolved 370 flaws , out of which 349 were reported by Google itself. Seven of the vulnerabilities have been marked critical in severity. The development comes amid an exponential surge in vulnerability discovery, mainly fueled by the advent of large language models (LLMs) that have accelerated the process, leading to an unprecedented spike in new bug reports, so much so that issues are being flagged at a faster rate than companies can fix them. According to statistics shared by the U.S. National Vulnerabilities Database (NVD), 46,872 flaws have been recorded so far in 2026, nearing the 49,920 vulnerabilities reported for the entirety of 2025.

An academic study has disclosed a "widespread class" of security vulnerabilities impacting 4G and 5G core networks that, if successfully exploited, could trigger denial-of-service (DoS) attacks and even session hijacking, allowing an attacker to seize control of a user's network session. The findings have been released by a group of researchers from Singapore's Nanyang Technological University in a paper titled "Understanding Implicit Trust Errors in Core Carrier Networks through Multi-Agent Flaw Discovery and Analysis." The study has uncovered dozens of vulnerabilities in the signaling interfaces of LTE/5G core networks, and specifically covers two LTE implementations (Open5GS and OpenAirInterface) and five 5G implementations (Open5GS, free5GC, OpenAirInterface, SD-Core, and eUPF) across two core signaling protocols, GPRS Tunnelling Protocol Control Plane ( GTP-C ) and Packet Forwarding Control Protocol ( PFCP ). "Our research finds these vulne...

Device code phishing - the abuse of the OAuth 2.0 device authorization grant to steal access tokens - has evolved from a niche red-team technique to an industrial-scale threat in under six months. Designed for input-constrained devices like smart TVs, printers, and so on, the device authorization login flow has been adopted by a wide range of apps and use-cases that it wasn't originally intended for - most commonly CLI logins. Researchers first described the attack vector in 2020, but it took until 2024 before nation-state actors like Storm-2372 started using it in the wild. By 2025, ShinyHunters was using device code phishing against Salesforce tenants at scale, then in February 2026, the EvilTokens kit arrived and criminal adoption skyrocketed. By April, Microsoft was reporting 10 to 15 entirely new campaigns every 24 hours . Barracuda counted 7 million attacks in four weeks . The FBI issued a standalone advisory on Kali365 , the first US federal agency PSA about a specif...

Palo Alto Networks' Unit 42 says a Chinese-speaking threat actor used DeepSeek through the open-source Hermes Agent framework to launch attacks autonomously. After an initial Telegram instruction, the agent found internet-facing systems and selected public exploits. The researchers recovered no further operator input in the session. The operator, tracked through the aliases knaithe and KnYuan , launched exploitation attempts against more than 460 targets using autonomous and conventional workflows. Unit 42 described seven exploit tracks. They span eight Common Vulnerabilities and Exposures (CVE) identifiers because the n8n chain combines two vulnerabilities. The DeepSeek-led attacks against Langflow and n8n failed because the exposed systems did not meet the exploits' configuration requirements. In separate manual operations, Unit 42 reported data exfiltration from three organizations through the NetScaler memory-overread flaw CVE-2026-3055 and command execution on 11...

Anthropic on Thursday became the latest artificial intelligence (AI) company to reveal that three of its models, including Claude Opus 4.7, Mythos 5, and an unnamed research model, had breached three unnamed organizations during cybersecurity testing without its knowledge. The AI firm said the earliest incidents date back to April 2026, adding it made the discoveries after launching a "large-scale retrospective review" in response to a recent disclosure from OpenAI that a combination of its models escaped the sandboxed environment by exploiting a previously unreported zero-day in Artifactory to obtain internet access and break into Hugging Face's production systems with an end goal to cheat on an evaluation. "After reviewing 141,006 evaluation runs where Claude could have obtained internet access, we identified three incidents in which a model accessed the internet from within or while interacting with the evaluation environment of Irregular, one of our third-...

A lot of security still comes down to trusting the wrong screen. This week, that screen might be a login page, an install guide, a recruiter call, or a familiar service behaving slightly wrong. Behind it: reused credentials, exposed systems, quiet loaders, abused trust, and exploit paths that should have been harder. Some defenses improved. The loose parts still got found first. Anyway, here's the mess. The threats change every week. Subscribe, and we’ll alert you when each new ThreatsDay Bulletin is out.

AI is compressing exploit timelines. The real question isn't whether your vulnerability management playbook needs to change, it's which part of it you've been getting wrong all along. The conversation happening in security circles right now goes something like this: Mythos is here. Exploit timelines are collapsing. Does the vulnerability management playbook need to change? The honest answer is yes. But not the part most people are focused on. The discussion around Mythos, Anthropic's frontier model and its implications for offensive security, tends to center on discovery. AI accelerates reconnaissance. It helps attackers identify exposures faster, chain techniques more efficiently, and move at machine speed through environments that were previously protected, in part, by the attacker's own time constraints. That's real. And it matters. But here's the part getting less attention: most security teams weren't winning the prioritization battle bef...

Most organizations have incident response plans, security tools, and technical teams in place. Yet new research suggests that many still lack the coordination, visibility, and executive alignment needed to withstand a serious cyberattack. According to The State of Incident Response Readiness 2026 , based on a survey of 600 senior IT security decision makers conducted by Vanson Bourne in January and February 2026, 73% of organizations admit they would not be "fully ready" if a significant cybersecurity attack occurred tomorrow. The findings point to a critical gap between having incident response capabilities and being able to execute them effectively under pressure. The report also found that cyberattacks are already a recurring business risk. More than three-quarters of organizations, 76%, experienced at least one cyberattack in the past 12 months, while 32% experienced more than one. Incident Response Readiness Remains a Weak Point Incident response has evolved f...

OpenAI on Tuesday revealed the rogue artificial intelligence (AI) agent that escaped its sealed evaluation environment and broke into Hugging Face's production environment also hacked multiple third-party accounts and services as part of the attack. The latest disclosure shows that the security incident, which stemmed from an internal security test, was more extensive in scope than previously thought. The AI company said its ongoing review of the incident revealed a "small number of cases" where the models, including GPT-5.6 Sol and an "even more capable pre-release model," identified and used exposed credentials at the account-level on other publicly-available services. "This includes four accounts on four services as part of the Hugging Face incident (and a few accounts accessed as part of other evaluations)," it said. "One of these four accounts was used as an outbound relay and staging path, and another account was used for data st...

Anthropic says Claude Mythos Preview helped derive an end-to-end key-recovery attack against HAWK-256 and a 200- to 800-fold speedup for an attack on seven-round AES-128. The HAWK attack exploits a previously unused symmetry in the lattice behind the signature scheme. Anthropic's released implementation gives an expected end-to-end runtime of about three hours and 42 minutes on a 96-core server. The AES result removes a 256-way guessing step from an existing meet-in-the-middle attack. Anthropic said neither result affects production systems. HAWK remains a candidate in a National Institute of Standards and Technology (NIST) post-quantum standardization process, and the public recovery code only targets the smaller HAWK-256 parameter. The Advanced Encryption Standard (AES) result applies to seven of AES-128's ten rounds and still requires an impractical number of chosen plaintexts. The company said no production software needs to change as a result. Anthropic published...

JFrog has confirmed that OpenAI models exploited a zero-day in self-hosted Artifactory while trying to reach the open internet from a sealed evaluation environment. Artifactory is JFrog's software repository manager. OpenAI says the models then escalated privileges and moved laterally until they reached an internet-connected node. JFrog says it has since developed and released fixes for cloud and self-hosted customers. The Artifactory exploit occurred inside OpenAI's environment. OpenAI says a separate attack path later reached Hugging Face's systems. JFrog says its cloud customers are already protected. Self-hosted users should review the Artifactory release notes and move to the remediating build for their maintained branch. Several Artifactory CVE records were published on July 27 with affected-version ranges and fixed-version thresholds, but neither JFrog nor OpenAI has said whether any of those records correspond to the vulnerabilities used during the evaluati...

OpenWrt has shipped version 24.10.8 to close a critical DHCPv6 stack overflow and a wider set of remotely triggerable flaws in network services enabled by default. The critical issue, tracked as CVE-2026-53921 and rated 9.8 on CVSS 3.1 in OpenWrt's GitHub advisory, lets an unauthenticated attacker able to reach the DHCPv6 server overwrite a stack buffer in odhcpd through a crafted DHCPv6 REQUEST. odhcpd runs as root, and the advisory notes that embedded hardware commonly lacks stack canaries and address space layout randomization (ASLR), making code execution a realistic outcome on typical devices. The advisory includes public Python proof-of-concept code for both documented overflow paths. Users on the 24.10 branch should install 24.10.8, while users on 25.12 should install 25.12.5; firmware images are available through the OpenWrt Firmware Selector. As of July 28, the reviewed OpenWrt materials did not report exploitation in the wild. The flaw was also absent from CISA...

STAR Labs has published a Linux kernel exploit that turns an ordinary local user into root on the CentOS Stream 9 build it targeted. The flaw, tracked as CVE-2026-53264 (CVSS score: 7.8), is a use-after-free race in the kernel's network traffic-control subsystem. Researcher Lee Jia Jie said artificial intelligence (AI) helped him find the bug and speed up exploit development. This is local privilege escalation, not remote code execution, so an attacker needs a foothold on the machine before any of it applies. The demonstrated exploit also requires unprivileged user namespaces, the CONFIG_NET_ACT_GACT and CONFIG_NET_CLS_FLOWER kernel options, and a kernel-specific return-oriented programming (ROP) chain containing hardcoded offsets. Those conditions narrow the immediate exposure, but full exploit source code is now public. The upstream fix landed on June 1, 2026, and has since been backported to several stable kernel branches. The Linux CNA record lists vulnerable ranges begi...

Microsoft has launched its first cybersecurity-specific model inside MDASH , its multi-model vulnerability identification and remediation harness. The company says MDASH, using MAI-Cyber-1-Flash and GPT-5.4, scored 95.95% on CyberGym. It also claims the configuration costs 50% less than its current best MDASH combination of GPT-5.4, GPT-5.4 mini, and GPT-5.3 Codex. Access is limited to approved MDASH customers through an Azure AI Foundry private preview. MAI-Cyber-1-Flash is designed to handle up to 90% of MDASH tasks, with GPT-5.4 reserved for the hardest 10%. It is available only inside MDASH, not as a standalone public model or general-purpose application programming interface. The headline score belongs to MDASH running MAI-Cyber-1-Flash alongside GPT-5.4, not to the new model by itself. CyberGym Level 1 is a known-vulnerability reproduction test. It gives an agent a vulnerability description and the corresponding unpatched source code, then checks whether it can produce a ...

Monday starts with the usual promise that everything is under control. Then the logs wake up. This week, trusted tools crossed lines, old flaws found new work, exposed systems stayed exposed, and attackers kept hiding inside normal-looking services. Nothing looked strange at first. That helped. That is the mood. Here is the full recap. ⚡ Threat of the Week OpenAI Says Its AI Agent Went Rogue and Targeted Hugging Face - OpenAI disclosed that it lost control of two AI models during a security evaluation that ended in a breach of Hugging Face. The AI company said its AI models broke out of a sealed testing environment and broke into Hugging Face's production system to find solutions for the ExploitGym benchmark. "The incident also makes clear that advanced models can discover and exploit novel attack paths in real-world systems without source-code access," OpenAI said. "It highlights that advanced cyber capabilities must be developed alongside stronger safegua...

The North Korean threat actors behind the ClickFix-style campaigns that employ typosquatted Zoom and Microsoft Teams domains have been found to operate an active phishing kit to impersonate the videoconferencing platforms in social engineering campaigns designed to deliver malware. " BlueNoroff has operationalised trust abuse by combining compromised industry contacts, social engineering, wallet reconnaissance and malware delivery into a repeatable victim acquisition pipeline," JUMPSEC said in a detailed report shared with The Hacker News. "The platform profiles victims' cryptocurrency wallets before malware delivery, enabling selective targeting of high-value victims." Describing the campaign as an operator-driven victim acquisition platform, the cybersecurity company noted that the activity involves using compromised trusted contacts as the initial access vector to create a self-propagating attack chain via Telegram. Details of the activity have been...

Someone installed a popular AI assistant on a rented server, switched off the setting that makes it ask permission before running risky commands, and pointed it at Thailand's Ministry of Finance, which runs the country's treasury and tax collection. The agent then worked through the ministry's network on its own, checking hosts for ways to gain root access, hunting through file systems, and crawling a folder of staff personnel records going back to 2012. The operator left the agent's own logs sitting on a web server with directory listing switched on, where threat intelligence firm Hunt.io and researcher Bob Diachenko found them, along with 585 files and 470 MB of attack tooling. The tool is Hermes , an open-source assistant from Nous Research that people install to manage their mail, run chores, and take instructions over Telegram or Slack. It is not a hacking tool, and nothing here is a flaw in it. The mode the operator used, called YOLO, is a documented feat...

Most of this week's trouble came dressed as something useful. A package stole data. A fake extension opened remote access. A safety app became spyware. An image gave hidden orders to an AI agent. Other threats hid in open systems, weak code, and normal network traffic. The threats change every week. Subscribe, and we’ll alert you when each new ThreatsDay Bulletin is out. The danger was easy to miss because it looked ordinary. Here is the full list:

Google on Thursday announced a new way for users to sign-in to their accounts by letting them take a selfie video. The selfie for sign-in , per the tech giant, is another option on top of existing recovery methods to log in to an account, including an email address or a phone number. The idea is to use a video selfie as a way to regain access if a user ever gets locked out or doesn't have access to their usual phone or computer. As part of the process, users are required to set up a selfie video by just looking into the device's camera and completing a "few short, guided head movements" to capture their face from different angles. Should users have any trouble signing in to their accounts with the selfie method at a later stage, they can just take another selfie to sign back in. "Selfie video compares the new video to the one you set up to confirm it is really you and help you get back into your account," Google said in a blog post shared with The Ha...

RefluXFS , a Linux kernel flaw disclosed on July 22 and tracked as CVE-2026-64600 , lets an unprivileged local user overwrite root-owned files on an XFS filesystem and gain persistent root access. Qualys said default installations of Red Hat Enterprise Linux and its derivatives, Fedora Server, and Amazon Linux can meet the conditions for exploitation. The company demonstrated the race against /etc/passwd and setuid-root binaries. The overwrite lands at the block layer. It survives a reboot and leaves the target's ownership, permissions, timestamps, and setuid bit untouched, so a modified setuid-root binary still runs as root. The fix was merged on July 16, and Linux vendors have begun shipping backported kernels. The patch traces the bug to Linux 4.11 in 2017: a Fixes: tag naming commit 3c68d44a2b49 and a stable backport request marked # v4.11 . Who is exposed Exploitation requires three conditions: The system runs Linux 4.11 or later without the RefluXFS fix....