The Hacker News | #1 Trusted Source for Cybersecurity News

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...

Credit: Hacktron Vercel has released security patches for two critical-severity vulnerabilities in the Next.js web framework, both of which allow unauthenticated remote code execution, one exploitable via specially crafted AVIF image files and the other through a path traversal flaw affecting servers that use a Windows filesystem. The Windows path traversal, tracked as  CVE-2026-75604  (CVSS score: 9.0), affects Next.js applications that use both the Pages Router and App Router without Cache Components when the server uses a Windows filesystem. Linux and macOS deployments are not affected. "There is no known workaround for affected windows-hosted applications. You should upgrade immediately if your server is hosted on Windows,"  Vercel said in its advisory . The fixes are available in Next.js 15.5.24 (Maintenance LTS) and 16.3.3 (Active LTS), published on August 25, 2026. Affected users can upgrade by running npm install next@15.5.24 for the 15.5 l...

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 Mindguard. 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," security researcher 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 ava...

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...

The Australian Federal Police (AFP) has charged two Western Australian men with a combined total of 14 offences over their alleged role in TeamPCP , the cybercrime group behind the March 2026 compromise of the open-source security scanners Trivy and Checkmarx KICS and the AI gateway LiteLLM. Louis Michael Gaebler , 23, and Ruben Ian Thomson , 21, appeared in Perth Magistrates Court on August 27, 2026, a day after the AFP and the Western Australia Police Force (WAPF) executed search warrants at properties in Cottesloe, Hamilton Hill, and Mandurah and seized electronic devices for forensic analysis. Police allege the two men were principal participants in the syndicate and received payments in cryptocurrency, the value of which is still under investigation. The Federal Bureau of Investigation (FBI) said in a July 2 advisory that organizations impacted by the campaign should treat exfiltrated data and credentials as a persistent risk, since affiliated threat actors are "likely...

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...

Individuals and organizations in Cambodia have emerged as the target of a new campaign that delivers an open-source remote access trojan (RAT) called Spark RAT . "The samples employ diverse lure themes, suggesting an effort to appeal to a broad range of potential victims. These include government notices, public health materials, real estate-related content, and other topics," Acronis Threat Research Unit (TRU) researchers Darrel Virtusio and Subhajeet Singha said in an analysis published Wednesday. The multi-stage attack is notable for employing the bring your own vulnerable driver (BYOVD) technique to load a legitimate-but-vulnerable driver associated with OPSWAT AppRemover ("ardrv.sys") to escalate privileges and neutralize security software. Attack chains likely make use of targeting phishing emails to distribute compressed archives containing an Inno Setup executable and trick recipients into running it using wide-ranging lures, including Cambodian gove...

Threat actors linked by Arctic Wolf to Dark Caracal with medium confidence deployed a previously undocumented Go-based malware framework, GoCaracal , during a June 2026 intrusion at an unnamed communications organization in Venezuela. GoCaracal provides operators with remote shell access and payload execution, while the extended profile adds browser data theft, keylogging, remote desktop control, and SOCKS5 proxying. Arctic Wolf also published a YARA rule and representative indicators of compromise (IoCs) that defenders can use to hunt for the malware. "We assess with medium confidence that this activity is linked to Dark Caracal," Arctic Wolf said. Arctic Wolf based the assessment on Bandook use, recurring Delphi-loader characteristics, Spanish-language financial lures, malicious SVGs, URL shorteners, document-themed infrastructure, hosting-provider preferences, and Latin American targeting. In its  technical analysis of GoCaracal , Arctic Wolf said the malware ap...

Academic researchers have disclosed a Rowhammer attack impacting NVIDIA workstation GPUs with GDDR6 memory that defeats error correction codes (ECC), the mitigation NVIDIA recommends against GPU Rowhammer, and enables denial-of-service (DoS) and privilege escalation to a root shell. Dubbed GPUThor , the attack was developed by researchers at the University of Toronto, who hammered four DRAM banks for 24 hours each on four Ampere-class cards, inducing bit flips on each. The following GPUs were tested and found vulnerable - RTX A6000 (48 GB GDDR6) RTX A5000 (24 GB GDDR6) RTX A4500 (20 GB GDDR6) RTX A4000 (16 GB GDDR6) Mounting the attack requires the ability to launch an unprivileged CUDA kernel on the target GPU, either as a co-tenant on a shared card or as untrusted code on a single-tenant machine. The researchers advise avoiding cross-tenant GPU sharing, monitoring ECC error counters, and restricting untrusted CUDA workloads. "Recently, researchers at the U...

The U.S. Cybersecurity and Infrastructure Security Agency (CISA) on Wednesday added six flaws to its Known Exploited Vulnerabilities ( KEV ) catalog, including a high-severity security vulnerability impacting Citrix NetScaler ADC and NetScaler Gateway, citing evidence of active exploitation. The vulnerabilities are listed below - CVE-2019-1068 - A remote code execution vulnerability in  Microsoft SQL Server that could allow an attacker to execute code in the context of the SQL Server Database Engine service account. CVE-2026-8452 - An improper restriction of operations within the bounds of a memory buffer vulnerability in Citrix NetScaler ADC and NetScaler Gateway that could lead to denial-of-service. CVE-2022-0995 - An out-of-bounds memory write vulnerability in  Linux Kernel that could allow a local user to gain privileged access or cause a denial of service on the system. CVE-2015-5287 - A privilege escalation vulnerability in  Red Hat Automatic B...

The U.S. Department of Justice (DoJ) on Wednesday announced the disruption of two hacking platforms named QScan and QTRouter operated by Chinese threat actors to target critical infrastructure and other sensitive networks in the country. The activity has been attributed to a Chinese state-sponsored group known as QTFY, employed by Nanjing Xinjiuwei Network Technology Company (南京鑫玖维网络科技有限公司).  "Among the victims of QTFY computer intrusion activity are the National Aeronautics and Space Administration, Federal Reserve, Department of Energy, Department of Justice, Department of Health and Human Services, National Institutes of Health, and the U.S. Senate," DoJ said. Damon Rouse, a security researcher at Lumen Black Lotus Labs who has been tracking the activity for over the past 18 months, told The Hacker News that the digital quartermaster has been active since May 2018. Nanjing counts both China's Ministry of State Security (MSS) and the People's Liberation Army...

Cybersecurity researchers have discovered additional infrastructure and previously undocumented malware associated with Nimbus Manticore , an Iranian state-sponsored hacking group affiliated with the Islamic Revolutionary Guard Corps (IRGC). Group-IB, in a new analysis published today, described the cyber espionage actor as among the most active Iranian APT groups in 2026. Nimbus Manticore (aka GalaxyGato, Mirage Kitten, Screening Serpens, Smoke Sandstorm, Subtle Snail, and UNC1549) is assessed to be linked to Tortoiseshell (aka Imperial Kitten and Unyielding Wasp), which is part of the Charming Kitten (aka Eclipsed Wasp) cluster. Tortoiseshell is known to be active since at least July 2018, mainly targeting defense, aerospace, IT service providers, and military organizations in the Middle East and the U.S. Nimbus Manticore also has a history of orchestrating its own version of the Dream Job campaign to deliver malware under the pretext of job opportunity-themed social engine...

Cybersecurity researchers have disclosed details of a new adversary-in-the-middle ( AitM ) phishing toolkit called NovaCookies that's used as a proxy to redirect Microsoft 365 sign-ins, while capturing authenticated sessions in the process. In a report shared with The Hacker News ahead of publication, Island characterized the $320/month service as a subscription-based phishing platform that facilitates real-time Microsoft 365 session theft. The kit has been used to target hundreds of organizations across multiple sectors in the U.S., the U.K., Canada, Germany, Israel, and the U.A.E. to date.  "Observed campaigns used genuine Docusign envelopes to carry counterfeit document-share lures, with some clicks routed through legitimate Microsoft or Google sign-in endpoints as redirect hops before reaching the kit," Island said. "The message, document service and redirect can therefore appear trustworthy until the browser reaches attacker-controlled infrastructure....

The U.S. Cybersecurity and Infrastructure Security Agency (CISA) has published the results of two red team assessments it conducted simultaneously against two critical infrastructure organizations, using what it described as similar tradecraft while recording sharply different defensive outcomes. Both organizations were fully compromised at the domain level, and in both, the red team also reached sensitive business systems (SBSs) and cloud resources. The advisory, tracked as AA26-237A and titled "A Tale of Two SOCs," was released on August 25, 2026. CISA identified the first target only as a Government Services and Facilities Sector organization, referred to as Organization A , and the second as a Water and Wastewater Systems Sector entity, referred to as Organization B . "CISA conducted two simultaneous red team assessments using similar tradecraft but observed different defensive responses," the agency said in the advisory. Against Organization A, the re...

The CERT Coordination Center (CERT/CC) has disclosed two unpatched vulnerabilities in Kaltura's HTML5 video player library that allow a remote, unauthenticated attacker to read arbitrary files from a server and execute code on it. The flaws, tracked as CVE-2026-19913 and CVE-2026-19912 , both stem from the same unsafe deserialization in the mwEmbedLoader.php endpoint of the mwEmbed player library, which Kaltura also distributes as html5lib. Neither requires authentication or a Kaltura session token, and network access to the endpoint is the only precondition CERT/CC states. No patch is available, and CERT/CC said it was "unable to reach Kaltura to coordinate these vulnerabilities." Administrators are advised to restrict or disable external access to the endpoint and to enforce a strict allow-list for the ServiceUrl parameter that permits only legitimate backend API URLs. No exploitation had been reported at the time of writing, and neither CVE appeared in CISA...