artificial intelligence | Breaking Cybersecurity News | The Hacker News

A major vulnerability is disclosed. The alert lands immediately. Then comes the harder question: Are we actually exposed? For many security teams, answering that means jumping between vulnerability scanners, endpoint tools, cloud inventories, SBOMs, repositories, and application data to build enough context to act. As AI accelerates vulnerability discovery and research, that delay matters more. The challenge is no longer just finding vulnerabilities. It is reducing the time between disclosure, exposure assessment, and response. Register now for our next live expert webinar, " Outpacing AI-Era Attacks Starts With Faster Time to Answer ," to see how the Tines security team approaches that problem. See How Tines Built a Faster Exposure View In this webinar, Tines Co-founder and CCO Thomas Kinsella and Senior Security Operations Engineer Andrew Katz will show how their security team brings multiple sources of exposure data into one interactive view.

A flaw in DeepSeek Harness , DeepSeek's open-source tool for running AI coding agents on a developer's machine, let a sandboxed agent turn off its own sandbox with a single command. The tool runs an agent's commands inside an operating-system sandbox, so that an agent working on untrusted files cannot write outside its workspace. The agent could remove that limit by calling the tool's own web interface on the same machine, and its commands would then run outside the sandbox without an approval prompt. It worked on a default installation until DeepSeek fixed the tool on August 27, and it required attacker-supplied text that the agent read to prompt it to make the call. The flaw is tracked as CVE-2026-82533 . VulnCheck, which assigned the identifier,  published the record  on September 8 and rated the flaw 9.4 out of 10. OX Research , the security firm that reported the flaw, said one shell command was enough. The command invoked the tool's local interface an...

U.S. cybersecurity and intelligence agencies have accused China-based artificial intelligence (AI) companies of conducting "systematic extraction" of proprietary functionalities and capabilities of American frontier models through distillation attacks. The activity has been described as occurring at an industrial-scale and one that forms the "core" of their AI development strategy, according to a bulletin released by the National Security Agency (NSA), the Cybersecurity and Infrastructure Security Agency (CISA), and the Federal Bureau of Investigation (FBI). "While 'distillation' is recognized as a legitimate and useful technique in AI research, China-based AI companies are engaging in aggressive, malicious, and targeted distillation activities at an industrial scale that extract restricted proprietary functionalities and capabilities of U.S. frontier AI models," the authoring agencies said . The joint advisory noted that Chinese AI firms l...

Google on Thursday released updates to patch 230 security vulnerabilities, including one that has come under active exploitation in the wild. The medium-severity vulnerability, assigned the CVE identifier CVE-2026-87491 (CVSS score: N/A), has been described as an out-of-bounds bug in V8, Chrome's JavaScript and WebAssembly engine. "Out-of-bounds write in V8 in Google Chrome prior to 153.0.8010.36 allowed a remote attacker to execute arbitrary code inside the sandbox via a crafted HTML page," reads a description of the flaw on the NIST National Vulnerability Database (NVD). Security researcher Jihyeon Jeong of Compsec Lab, Seoul National University, has been acknowledged for discovering and reporting the flaw on August 6, 2026. The researcher received a $2,500 bug bounty reward for responsible disclosure. Google acknowledged it is "aware that an exploit for CVE-2026-87491 exists in the wild," but has not disclosed any additional specific information r...

Check Point Research said in a report published today that a single instruction planted in a ChatGPT conversation could cause ChatGPT to quietly work for an attacker while answering the user's question as usual. In the company's proof of concept, that hidden work read data from the user's connected Gmail account and passed it to a second ChatGPT account through a hidden channel between the two. The reply the user saw said nothing about it. Check Point said the same channel could also copy out the chat history and the files in that conversation. How much an attacker could take depended on what the session could already access, including its data, tools, other connected apps, and permissions. The instruction had to be in the conversation before any of this worked. Check Point named three ways to get it there: a prompt the user pastes in, a shared ChatGPT conversation the user opens, or a custom GPT that holds it in its builder instructions, which are not shown to th...

Threat actors are continuing to leverage artificial intelligence (AI) to streamline their operations , with one financially motivated hacking group employing an autonomous, multi-agent attack framework to carry out a large-scale credential harvesting campaign within six hours. Google Threat Intelligence Group (GTIG) said it has observed attackers with diverse motivations targeting proprietary AI models across healthcare, government, and media sectors, exfiltrating API credentials, and co-opting victim cloud environments to sustain unauthorized AI workloads. This highlights growing attacker focus on enterprise AI assets for espionage, extortion, and resource theft. "At this point, we can assume that all threat actors are using AI in some capacity and their operations have benefited," John Hultquist, chief analyst at GTIG, said in a statement shared with The Hacker News. "Like everyone else, we're concerned about the vulnerability problem, but AI is being applie...

Researchers at the security firm Calif have built a worm that takes over a WeChat account via an incoming call and  demonstrated it spreading  among three test phones. The person being called does not have to answer or touch their phone for it to work, but the caller must already be one of their WeChat contacts. Calif reported the flaw to Tencent in July and says the company has since blocked the exploit for all users. No attacks using the flaw have been reported, and Calif does not say there were any. Attacks that require no action from the target, known as zero-click attacks, are not new. Last year, WhatsApp  patched a flaw  it said may have been used in targeted attacks. Answering the call does not stop the attack. Calif said a person who picks up hears nothing and the exploit still works. Declining the call ends that attempt, but the attacker can call again later, for example while the target is asleep. The caller has to be on the target's WeChat contac...

In the last six months, Chainguard doubled its output from 500 million to more than 1 billion container build manifests. We also surpassed 3,000 unique container images and 675,000 image versions in our catalog. Those are the headline numbers, but I want to share what's actually behind them. The number itself is less interesting than the system that produced it, and why we had to fundamentally rethink that system to get here. What a build manifest actually represents Let’s be precise about what we're counting. How do we define a “build manifest”? Think of it as every time the Chainguard Factory produces a new, verifiable artifact: a fresh image for go:1.26.5 , a rebuild of nginx triggered by a libc patch, a new architecture variant, a regenerated SBOM after a dependency change - all these events cause a new build and therefore new artifacts. At our scale, a single project like Python might have dozens of supported versions, each with multiple architecture builds, each r...

Turning off email images should at least stop the pictures. This week, attackers had a workaround: a scannable QR code built out of text. It still appears, even with images blocked. A small detail, but an annoying one if that was a precaution you were counting on. Elsewhere, a trusted software source delivered code that stole credentials, and a protocol designed for secure network management gave outsiders useful clues before login. Add active attacks on browsers, routers, and online stores, and there’s plenty to check—even for teams that have kept up with the patches. Read the full recap for the week’s major developments, plus more research, attacks, and security news beyond what we covered last week. ⚡ Threat of the Week N-able Patches Critical N-central Flaws — N-able has released hotfixes to address two severe N-central flaws (CVE-2026-86206 and CVE-2026-86207) that could allow an unauthorized party to bypass authentication controls and gain full access to the platform. ...

A group of AI safety researchers says a fleet of autonomous agents that identified themselves as OpenAI systems left about 18,000 posts on a dormant 25-year-old German wiki between May and July 2026, using the site as a shared board to pool answers to a timed web task and pass around a way out of their sandbox. The activity was concentrated on DSEwiki , a German software developer wiki that runs on the ProWiki farm at wikiservice[.]at and had been edited about 20 times over the previous decade. The researchers, led by Sydney Von Arx of the AI safety nonprofit  Nightingale Collective , reconstructed the deleted pages from edit history and published their analysis along with a downloadable copy of the data. The wikis let anyone change a page with an ordinary web request, the kind normally used only to read a page rather than to write one. Agents that were restricted to reading the internet could therefore still write to the site, because the restriction was written a...

Microsoft is alerting of a "high-volume phishing campaign" that's using invisible Unicode tag characters to bypass email filters. "Instead of using these characters to hide instructions from people while exposing them to AI models, the attacker used them to split financial lure words such as 'funding' to prevent email filters from parsing them," the Microsoft Security Research team said . The Windows maker said the findings show AI-era evasion techniques can be adapted by threat actors in traditional phishing and spam campaigns. Attacks exploiting this approach are said to have first emerged in early February 2026. ASCII Smuggling refers to a technique where invisible or non-rendering Unicode characters are used to conceal messages or instructions inside seemingly-harmless text. As a result, human user interfaces do not render them, making the text appear completely normal to the user. However, such content can be ingested by email filters or AI...

OpenAI on Thursday officially unveiled GPT‑6 Astra , which it described as the "world's most intelligent and aligned model." The development comes days after the artificial intelligence (AI) company said the model had reached the "Critical" cybersecurity capability threshold under its Preparedness Framework. "Astra is state-of-the-art on computer use, browsing, software engineering, cybersecurity, science, and professional work. Astra saturates FrontierMath Tier 4 with a 98% score," OpenAI said. "Astra also saturates ARC-AGI-3 with a 99.9% score and ExploitBench with a 100% score. It also sets a new frontier on computer and browser use, handling the most demanding professional work with unmatched speed, accuracy, and judgment." The model is currently rolling out to a small set of organizations and is expected to be available to all ChatGPT Plus, Pro, Business, and Enterprise users, as well as through the OpenAI API, Microsoft Azure,...

The worst part is how normal these attacks look. A call from IT. A shared file. A trusted app. A simple request to click “Allow.” Why break in when someone might open the door? That idea runs through this edition. Attackers use real tools, fake login pages, old account links, and software guides that point to unsafe downloads. One wrong letter in a web address can be enough. There is also ransomware, stolen ID data, hidden attack servers, and weak settings that should have been fixed long ago. Here’s the full list. 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 sophisticated Python-based Windows malware framework called BraZetsu that fuels an underground marketplace commercializing access to compromised hosts. "Unlike the standard infostealer model, BraZetsu is a comprehensive master toolkit that empowers Initial Access Brokers (IABs) by turning compromised systems into highly valuable commercial assets," Group-IB malware analysts Julio Guapo Menezes and Miguel Salazar said in a technical report. "The framework exhibits high operational maturity, utilizing a modular architecture and stealth techniques that allowed some samples to remain fully undetectable on VirusTotal at the time of analysis." BraZetsu is a portmanteau of "Brazil" and "Zetsu," a fictional character from the Japanese Manga series Naruto who is known to operate as a threat from the shadows. The naming is inspired by the fact that the initial access tool stealthily infiltrat...

The U.S. Cybersecurity and Infrastructure Security Agency (CISA) on Wednesday added seven security flaws to its Known Exploited Vulnerabilities ( KEV ) catalog after they landed in attackers' crosshairs. The vulnerabilities are as follows - CVE-2026-83548 (CVSS score: 10.0) - A server-side request forgery vulnerability in SonicWall SMA 1000 Appliances that could allow a remote unauthenticated attacker to gain unauthorized access to sensitive functionality and perform unauthorized operations.  CVE-2026-83549 (CVSS score: 7.8) - A post-authentication operating system command injection vulnerability in SonicWall SMA 1000 Appliances that could enable a remote authenticated attacker as administrator to execute arbitrary OS commands, resulting in remote code execution. CVE-2026-9586 (CVSS score: 9.3) - An SQL injection vulnerability in Sangoma Switchvox that could allow an unauthenticated remote attacker to execute arbitrary SQL statements against the backend PostgreSQL...

Google on Wednesday announced Gemini 3.8 Flash Cyber, which it described as its most capable cybersecurity model, and has made it available to a set of trusted defenders via a new initiative called the Fairwind Program . "The Fairwind Program gives high-priority defenders (like governments, healthcare providers, and telecommunications services) early access to advanced models that help them build better defenses, before new threats arrive," Google said . "So defenders have an early advantage, to help them protect vital infrastructure – which in turn protects people who rely on those systems." The tech giant said it's currently working with over 650 partners globally, including CrowdStrike, Datadog, Menlo Security, Palo Alto Networks, and Snowflake. The program is available to a group of Google Cloud customers, government agencies, and cybersecurity partners. The release of Gemini 3.8 Flash Cyber comes a little over a month after Google unveiled Gemini 3....

Manifold Security has disclosed eight security flaws across seven command-line AI coding agents in which a repository's own Git configuration names a command that the agent runs on the developer's machine, four of them still unpatched at publication. The command executes as the user, outside the agent's sandbox and without an approval prompt, and exploitation requires the repository to arrive as files with its .git directory intact, which a shared archive, a shared drive, a sync folder, or a USB stick preserves, whereas an ordinary clone does not. Fixes have shipped for goose, Claude Code, and Cursor, while Hermes Agent, Qwen Code, Grok Build, and a second path in Claude Code were still executing repository-supplied commands when Manifold retested them on September 1. OpenAI published three CVEs of its own the same day covering the identical class in Codex, credited to three unrelated research groups. "The helper runs outside Codex's command sandbox and wi...

The debate about whether AI delivers business value is over. The challenge now is implementing it at scale and securely across every function while meeting board-level pressure to move fast. Organizations must focus on adopting AI at business speed without losing control of cyber risk. Download the full eBook here.   The Business Reality   In Sygnia’s 2026 CISO Survey Report , which surveyed 600 senior IT and security leaders worldwide, nearly one-third already report extensive AI use across threat detection and IR, with 63% expecting it to be fully embedded in their organization by 2027. 1 Yet 73% of IT security decision makers say their organization would not be fully ready if a significant cyberattack occurred tomorrow. 1 Security teams feel they do not have adequate time to adapt. The tools are being deployed. The governance, controls, and incident readiness to support them are not. Security leaders are now tasked with enabling AI adoption while reducing the inheri...

Forescout Research - Vedere Labs said it used Anthropic's Claude to port a working pre-authentication remote code execution (RCE) exploit from one WAGO programmable logic controller (PLC) to another, executing attacker-supplied ARM shellcode on live hardware. The exploit targets  CVE-2021-31886 , a stack-based buffer overflow in the Nucleus FTP server's handling of the USER command, which carries a Siemens-assigned CVSS score of 9.8 and is accessible before authentication over TCP port 21. CERT@VDE says no updates are available for the affected WAGO controllers, and advises owners to disable or block FTP on port 21, enforce segmentation controls, and monitor network traffic for anomalies. The port required sustained researcher steering, and the final RCE development stage consumed $535.74 in application programming interface (API) usage over an 8-hour-32-minute session. A later session that attempted to extend the exploit into a command-and-control (C2) implant w...

Brazilian financial services, retail, and e-commerce organizations have become the target of a financially motivated threat actor dubbed Breeze Comet (formerly UNC5669) since 2024. Google Threat Intelligence Group (GTIG) and Mandiant teams described the threat actor as "specializing in manipulating payment systems and banking software in Brazil to conduct fraudulent transfers." The adversary is said to have successfully carried out at least one heist of assets worth tens of thousands of U.S. dollars. The activity overlaps with threat activity clusters tracked by Palo Alto Networks Unit 42, CrowdStrike and Trend Micro under the monikers CL-CRI-1163, Plump Spider and SHADOW-AETHER-064 . According to CrowdStrike, the e-crime group is operating out of Brazil and has been active since September 2023, monetizing their intrusions by gaining unauthorized access to internal payment systems and carrying out fraudulent transactions. Initial access to financial entities and com...