Cybersecurity Research | Breaking Cybersecurity News | The Hacker News

OpenAI on Wednesday detailed a pipeline that employed an upcoming cyber-focused artificial intelligence (AI) model to validate and patch 948 local privilege escalation vulnerabilities in Linux for $1.2 million in API costs. "Our prototype identified and analyzed 2,308 candidate Linux kernel findings, dynamically reproduced 1,724 vulnerabilities, and produced 948 local privilege escalation (LPE) exploits that could allow an unprivileged local user to gain root access," the AI company said . The findings build upon 24 LPE exploits that OpenAI revealed were discovered by its GPT‑5.5‑Cyber model back in June 2026. The latest research turns this process into an agentic pipeline comprising five stages - Vulnerability discovery, where agents identify high-value audit targets (i.e., parts of the codebase that are security-relevant and reachable by low-privileged users) and look for candidate findings Dynamic validation, where agents attempt to reproduce each candidate ...

Cybersecurity researchers have disclosed details of an OpenAI Codex jailbreak bundle that's designed to modify an artificial intelligence (AI) agent's trusted instructions to insert advertisements for a third-party AI gateway. The phenomenon has been codenamed prompt adware by Check Point Research. Called Haiou (meaning Seagull in Chinese), the jailbreak modifies OpenAI Codex's configuration files to inject advertisements for the jailbreak author's website into subsequent agent responses. "The advertised service is an opaque, OpenAI-compatible LLM gateway, a middleware between an application and an LLM provider," Check Point Research said in a technical report. "The likely business model is to use the freely distributed jailbreak bundle as an acquisition channel for the paid gateway service." Adware is a type of software that displays unwanted advertisements, which can appear on users' computers or mobile devices, often in the form of ...

Right now, there is probably an AI agent running on a laptop somewhere in your network. It stores API keys and database passwords, can access your internal systems, and works without anyone watching. Nobody reviewed its access. Nobody owns it. If an employee had that much reach with that little oversight, you would call it an insider risk. So why do we still call the agent a productivity tool? Most of the conversation about AI security is about what threat actors are doing with it. But the bigger threat may already be inside your organization.

Cybersecurity researchers have disclosed details of a  remote Spectre attack  against Cloudflare Workers that leaked a JSON Web Token (JWT) from a co-located Worker in the production environment at up to 12 bits per second, 360 times the rate of an earlier attack demonstrated in 2021. The end-to-end experiment used an attacker Worker and a victim Worker controlled by the researchers, with the JWT intentionally placed in the victim's memory. The research paper stated that no customer data was accessed. Cloudflare said the attack has already been mitigated in production after it improved Dynamic Process Isolation (DyPrIs), integrated the  V8 Sandbox , and deployed Memory Protection Keys (MPK)-based in-process isolation, adding that it found no indicators of active exploitation over the last three years. "We demonstrate that the production implementation of DyPrIs was insufficient," the researchers said in  the paper . Cloudflare Workers runs code from multiple...

Nebula Security says a patched Firefox JIT flaw could be triggered by simply visiting a malicious webpage and was also used to compromise Tor Browser. Tracked as CVE-2026-10702 , the bug provides arbitrary code execution inside the browser's renderer process. Mozilla rated it High and fixed it in the Firefox 151.0.3 update . "No settings or additional user interaction are required," Eten Zou, CEO of Nebula Security, told The Hacker News. "Visiting a malicious webpage is enough to trigger it," Zou said every Tor Browser release that incorporated a vulnerable Firefox version was affected, though researchers have not identified the exact Tor releases. On its own, the bug runs code only inside Firefox's sandboxed content process. Nebula released public exploit material and used the flaw as the first stage of IonStack, a browser-to-kernel chain built for an ARM64 device running Android 17. The released end-to-end code targets one supported Google build, a...

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...

An AI coding assistant that refuses to answer a dangerous request in its chat box can answer it anyway if the same request is broken into small, ordinary-looking steps inside a code editor. That is the finding of a  new study of GitHub Copilot  by researchers Abhishek Kumar and Carsten Maple. The models they tested through Copilot, Claude from Anthropic, and Gemini from Google, refused almost every harmful request when asked directly. Reframed as steps in a normal coding task, they produced the harmful answers in all 816 of the study's workflow runs. What makes this different from a typical jailbreak: no one asks for the harmful thing directly, and the model is not tricked into running someone else's code. It writes the banned content itself, as a side effect of a coding task it was told to improve. How it works The researchers call the method workflow-level jailbreak construction . Instead of a single blunt prompt, they asked Copilot to build an everyday piece of s...

Last year, there were discussions about Google Code—a platform that lets developers host their projects—being exploited to distribute malware. Research by zScaler has identified yet another instance where this platform has been misused. According to the Google Code site: "Project Hosting on Google Code provides a free collaborative development environment for open source projects. Each project includes its own member controls, Subversion/Mercurial repository, issue tracker, wiki pages, and downloads section. Our hosting service is designed to be simple, fast, reliable, and scalable, enabling you to concentrate on your open source development." The concerning project contained over 50 executable files in its download section. These files, mainly executable and zipped ".rar" files, have been uploaded over the past month, indicating that an attacker is actively using this free service to disseminate malware. VirusTotal results for the first file revealed that only 8 ...