IA Descubre Fallo Critico en Linux Como Protegerte

La inteligencia artificial (IA) se está posicionando como una herramienta revolucionaria en el campo de la ciberseguridad, y no solo para la defensa, sino también para el descubrimiento de vulnerabilidades. Un reciente hallazgo pone de manifiesto cómo la IA ha sido fundamental en la identificación de una falla de seguridad crítica en el kernel de Linux, un exploit de tipo 'race condition' que podría permitir a un atacante obtener privilegios de root.

Este descubrimiento subraya la creciente sofisticación de las amenazas cibernéticas y la necesidad imperante de que las organizac iones fortalezcan sus defensas. La investigación ha revelado que un modelo de IA, diseñado para detectar patrones de código complejos y comportamientos anómalos, fue clave para identificar esta vulnerabilidad específica que había pasado desapercibida para los métodos de detección tradicionales. El exploit, conocido como 'race condition', explota las debilidades inherentes a la concurrencia en el código, permitiendo a un atacante manipular el orden de ejecución de ciertas operaciones para obtener un acceso no autorizado y elevados privilegios dentro del sistema.

Ante esta nueva realidad, es crucial que las empresas adopten un enfoque proactivo para la seguridad. La recomendación principal es implementar una estrategia de defensa en profundidad que incluya la monitorización continua de la red, la aplicación inmediata de parches de seguridad, y la realización de auditorías de seguridad regulares. Además, es fundamental capacitar al personal en ciberseguridad para que reconozcan y reporten actividades sospechosas. La IA, si bien puede ser una herramienta para descubrir fallos, también puede ser empleada para fortalecer las defensas, mediante el análisis de logs, la detección de anomalías en tiempo real y la automatización de respuestas a incidentes. Por lo tanto, la integración de soluciones de seguridad basadas en IA se vuelve cada vez más relevante para mantenerse un paso adelante de las amenazas emergentes y garantizar la protección de los datos y sistemas críticos.

Fuente Original: https://thehackernews.com/2026/07/researcher-says-ai-helped-develop-linux.html

Artículos relacionados de LaRebelión:

• IA Hackea Finanzas Tailandesas Como Protegerte
• Rust Transforma Linux Programar es Divertido Nuevamente
• Linus Torvalds Defends AI Coding in Linux
• Linus Torvalds Defends AI Coding in Linux
• Pagos AI Linux Foundation Transforma el Futuro Financiero

Artículo generado mediante LaRebelionBOT

Rust Transforma Linux Programar es Divertido Nuevamente

Greg Kroah-Hartman, mantenedor del kernel estable de Linux, ha pasado de ser escéptico a convertirse en uno de los defensores más firmes de Rust dentro del núcleo de Linux. En su presentación durante la Open Source Summit India 2026 en Mumbai, Kroah-Hartman declaró que Rust ya no es un experimento, sino una parte permanente del ecosistema Linux que está revolucionando la forma en que se desarrolla el kernel.

Inicialmente resistente al cambio, Kroah-Hartman confesó que cuando un amigo le sugirió probar Rust hace años, él insistió en que C era suficiente. Sin embargo, su amigo tenía razón: Rust hace que la programación vuelva a ser divertida. Según sus palabras, el lenguaje elimina muchas preocupaciones relacionadas con el compilador, que puede resolver problemas automáticamente, mejorando significativamente la calidad del código.

La razón principal detrás de este entusiasmo es la capacidad de Rust para eliminar vulnerabilidades. Linux enfrenta aproximadamente 13 CVEs (Vulnerabilidades y Exposiciones Comunes) diariamente, con cambios que ocurren a un ritmo de casi nueve por hora durante más de una década. Kroah-Hartman estima que el 80% de estas vulnerabilidades desaparecerían simplemente porque serían detectadas por el sistema de tipos y propiedad de Rust. La mayoría de estos problemas no son ataques sofisticados, sino errores simples de C: punteros sin verificar, bloqueos olvidados y rutas de limpieza descuidadas.

El impacto de Rust ya es tangible en el kernel. Algunos subsistemas solo aceptarán nuevos controladores escritos en Rust. Un ejemplo destacado es Binder, el mecanismo de comunicación entre procesos de Android presente en miles de millones de dispositivos. Actualmente existen implementaciones paralelas en C y Rust, pero la versión en C pronto desaparecerá, dejando a Rust como la base de todos los dispositivos Android en el futuro. Esta transición refleja la confianza creciente en Rust para mejorar la seguridad y eficiencia del kernel Linux.

Fuente Original: https://developers.slashdot.org/story/26/07/20/0417244/rust-will-help-linux-succeed-and-makes-coding-fun-says-greg-kroah-hartman?utm_source=rss1.0mainlinkanon&utm_medium=feed

Artículos relacionados de LaRebelión:

• Linus Torvalds Defends AI Coding in Linux
• Linus Torvalds Defends AI Coding in Linux
• Pagos AI Linux Foundation Transforma el Futuro Financiero
• Linus Torvalds Defiende C sobre Rust y IA
• Google Transfers Open Health Stack to Linux Foundation

Artículo generado mediante LaRebelionBOT

Linus Torvalds Defends AI Coding in Linux

Linus Torvalds has made it abundantly clear that the Linux kernel will not be banning AI-assisted coding tools, delivering a blunt message to critics: if you disagree with this decision, you're welcome to fork the project or simply walk away. In a characteristically direct post on the Linux kernel mailing list, Torvalds firmly stated that Linux is not an anti-AI project and will continue to embrace these emerging technologies based on technical merit rather than fear.

The controversy erupted following discussions about Sashiko, an agentic Linux kernel code review system that uses AI to identify bugs. According to its creators, the tool can independently detect approximately 53.6 per cent of bugs that human coders would eventually fix in later commits. However, the system also generates false positives—incorrectly flagging non-existent bugs—at a rate estimated to be within the 20 per cent range. This has sparked debate amongst maintainers about whether the benefits of automated bug detection outweigh the time cost of reviewing erroneous reports.

The discussion intensified when participants referenced the Software Freedom Conservancy's recent statement advocating for the rights of open source contributors to reject LLM-generated AI systems entirely. The Conservancy argued that the community should actively support, not merely tolerate, those who choose to avoid AI tools altogether. Torvalds, however, firmly rejected this absolutist position, stating that whilst no one is being forced to use AI tools, he will "very loudly ignore" those who attempt to prevent others from utilising them.

Torvalds defended his stance as pragmatic and grounded in technical merit, asserting that AI has clearly proven itself as a useful tool. He suggested that anyone doubting AI's utility simply hasn't used it properly, noting that whilst the technology may not have been clearly beneficial even a year ago, its value is no longer in question today. Responding to concerns about AI imperfections, Torvalds urged critics to examine their own performance, pointing out that natural intelligence isn't always superior either. His message is clear: AI is simply another tool in the developer's toolkit, and Linux will continue to welcome its use.

Fuente Original: https://linux.slashdot.org/story/26/07/17/1830258/linus-torvalds-to-critics-of-ai-coding-on-linux-fork-it-or-just-walk-away?utm_source=rss1.0mainlinkanon&utm_medium=feed

Artículos relacionados de LaRebelión:

• Pagos AI Linux Foundation Transforma el Futuro Financiero
• Linus Torvalds Defiende C sobre Rust y IA
• Linus Torvalds AI Coding Tools Reality Check
• Google Transfers Open Health Stack to Linux Foundation
• Google Awards 250K for Critical Linux VM Escape

Artículo generado mediante LaRebelionBOT

Linus Torvalds Defends AI Coding in Linux

The debate surrounding artificial intelligence in software development has reached a critical juncture, with Linux creator Linus Torvalds taking a firm stance in favour of AI-powered coding tools. In a recent mailing list post, Torvalds made it abundantly clear that Linux will not join the ranks of anti-AI projects, telling critics they are free to fork the project or simply walk away if they disagree with this direction.

The controversy centres around Sashiko, an AI-driven code review system designed specifically for the Linux kernel. According to its developers, the tool can independently identify approximately 53.6 percent of bugs that human programmers would eventually fix in subsequent commits. However, the system also generates false positives at an estimated rate of around 20 percent, potentially creating additional work for maintainers who must sift through erroneous bug reports.

Torvalds rejected arguments from those citing the Software Freedom Conservancy's recent statement supporting contributors who refuse to work with AI-generated code. Whilst acknowledging that no one is being forced to use these tools, he stated he would "very loudly ignore" those attempting to prevent others from utilising them. His position, he insists, is rooted in technical merit rather than fear of innovation, describing AI as simply another useful tool in the developer's arsenal.

The Linux founder's pragmatic approach is informed by his own experimentation with AI coding tools. In January, Torvalds shared his experience using what he termed "vibe coding" tools to create a Python audio visualiser for a guitar pedal project, noting how AI streamlined the development process. He argues that whilst AI isn't perfect, human intelligence has its own significant limitations, and critics should be equally self-aware about their own fallibilities.

Research on AI coding productivity presents mixed findings. A 2025 study found developers using AI tools were actually 19 percent less productive despite feeling more efficient. However, updated research from early 2026 suggests genuine productivity gains are now materialising as the technology matures. Not everyone embraces this shift—some developers have even embedded malicious prompt-injection instructions to sabotage AI tools attempting to modify their code.

Fuente Original: https://arstechnica.com/ai/2026/07/linus-torvalds-to-critics-of-ai-coding-in-linux-fork-it-or-just-walk-away/

Artículos relacionados de LaRebelión:

• Pagos AI Linux Foundation Transforma el Futuro Financiero
• Linus Torvalds Defiende C sobre Rust y IA
• Linus Torvalds AI Coding Tools Reality Check
• Google Transfers Open Health Stack to Linux Foundation
• Google Awards 250K for Critical Linux VM Escape

Artículo generado mediante LaRebelionBOT

Pagos AI Linux Foundation Transforma el Futuro Financiero

La Linux Foundation ha dado un paso audaz hacia la integración de la inteligencia artificial en el tejido económico de internet. Con el lanzamiento de la Fundación x402, se busca establecer un estándar para los pagos nativos de internet, diseñados específicamente para agentes de IA, APIs y aplicaciones. Este ambicioso proyecto se fundamenta en el protocolo x402, donado por Coinbase, y cuenta con el respaldo de gigantes tecnológicos y financieros como AWS, American Express, Cloudflare, Google, Mastercard, Stripe y Visa.

La visión detrás de x402 es permitir que los agentes de IA realicen transacciones financieras directamente a través de HTTP, abriendo un abanico de nuevas posibilidades. Lincoln Murr, líder de productos de IA en Coinbase, destaca que la meta es d otar a estos agentes de independencia financiera, ampliando drásticamente sus capacidades para interactuar con cualquier servicio en línea. La analogía se traza con la transición de la era del escritorio a la móvil en la década de 2010; ahora, se vislumbra que los agentes de IA se convertirán en los actores económicos predominantes en la internet de finales de la década de 2020.

Esta iniciativa promete revolucionar cómo operan las aplicaciones y servicios en línea, al permitir que la IA gestione pagos de manera autónoma y eficiente. La estandarización de estos pagos es crucial para asegurar la interoperabilidad y la seguridad en un ecosistema cada vez más impulsado por la inteligencia artificial. La implicación para los usuarios, aunque requiere una consideración sobre la comodidad de delegar transacciones financieras a la IA, abre la puerta a experiencias más fluidas y a la automatización de procesos complejos que antes requerían intervención humana.

Fuente Original: https://linux.slashdot.org/story/26/07/14/1853255/linux-foundations-latest-foray-is-to-standardize-internet-native-payments-for-ai-agents?utm_source=rss1.0mainlinkanon&utm_medium=feed

Artículos relacionados de LaRebelión:

• Nueva York detiene centros de datos El futuro de la IA
• IA China Copia Modelos Americanos Se Roba el Futuro
• Google Transfers Open Health Stack to Linux Foundation
• Google Awards 250K for Critical Linux VM Escape
• GhostLock Flaw 15-Year Linux Vulnerability Exposed

Artículo generado mediante LaRebelionBOT

Google Transfers Open Health Stack to Linux Foundation

The Linux Foundation is set to launch the Open Health Stack Software Foundation, a new vendor-neutral organisation designed to house Google's Open Health Stack project. This significant move sees Google contributing all project code and assets, whilst Google.org provides a substantial $3 million grant to support the initiative. The project has garnered backing from major technology players including Microsoft and Anthropic, alongside the World Health Organisation, all united in their mission to develop open source, AI-ready digital health infrastructure accessible to healthcare systems worldwide.

At the heart of this initiative lies a fundamental question: will transferring governance to the Linux Foundation genuinely accelerate adoption across the healthcare sector, or will this become yet another foundation that remains largely disconnected from the average developer's daily work? The project's focus centres on core HL7 FHIR technologies, which serve as the backbone for healthcare interoperability, enabling different medical systems to communicate effectively. Additionally, the Open Health Stack Player deployment toolkit aims to simplify implementation, whilst AI Commons represents a model-agnostic healthcare artificial intelligence initiative being co-developed with the World Health Organisation.

What distinguishes this announcement from typical technology foundation launches is its planned Implementer Programme, which represents a deliberate effort to democratise healthcare technology development. This programme specifically aims to provide startups, small businesses, and local developers in low- and middle-income countries with a formal role in project governance. Rather than simply establishing healthcare software standards from the top down, the initiative recognises the critical importance of ensuring that those implementing these technologies in underserved markets have a genuine voice in shaping the project's direction. This approach acknowledges that healthcare challenges vary dramatically across different economic contexts, and solutions must be informed by the experiences of those working on the ground in diverse healthcare environments worldwide.

Fuente Original: https://news.slashdot.org/story/26/07/09/1659227/google-hands-open-health-stack-to-the-linux-foundation?utm_source=rss1.0mainlinkanon&utm_medium=feed

Artículos relacionados de LaRebelión:

• Google Awards 250K for Critical Linux VM Escape
• GhostLock Flaw 15-Year Linux Vulnerability Exposed
• Bad Epoll Flaw Grants Root Access on Linux
• Microsoft Unveils Linux Containers Running on Windows
• Vulnerabilidades AI Linux y Malware Informate Ya

Artículo generado mediante LaRebelionBOT

Google Awards 250K for Critical Linux VM Escape

Two significant security vulnerabilities have emerged in the Linux operating system this week, with Google awarding substantial bounties for their discovery. The more severe of the two, dubbed Januscape, allows untrusted virtual machines to break free from their isolated environments and gain root access to host machines—a nightmare scenario for cloud computing platforms.

The Januscape vulnerability, tracked as CVE-2026-53359, affects the Kernel-based Virtual Machine (KVM) component found in many Linux distributions. What makes this flaw particularly concerning is that it remained undetected for 16 years, silently lurking in code used by countless cloud providers worldwide. The vulnerability works on both AMD and Intel processors, exploiting bugs in the guest-side portion of virtual machines. Security researcher Hyunwoo Kim discovered that an attacker renting a single cloud instance could potentially crash the entire host system, taking down all other tenant virtual machines, or worse—execute code with root privileges to completely compromise the host and every guest running on it.

The technical nature of Januscape involves a use-after-free vulnerability in the shadow MMU emulation process, which handles memory address translation between hosts and hypervisors. Kim has released a proof-of-concept exploit that demonstrates a host crash, whilst keeping a full escape exploit under wraps for the foreseeable future. Google recognised the severity of this discovery with a $250,000 bounty through their kernelCTF programme.

The second vulnerability, named GhostLock and tracked as CVE-2026-43499, presents a different but equally serious threat. This flaw allows users with limited privileges to escalate their access to root level. Discovered by researchers at Nebula Security using AI-assisted scanning, GhostLock had been hiding in the Linux kernel for 15 years. The vulnerability resides in the futex priority-inheritance system, where a cleanup operation executes at the wrong moment, creating another use-after-free condition. With a severity rating of 7.8 out of 10, this flaw earned Nebula Security a $92,337 bounty from Google.

Both vulnerabilities have now been patched in the Linux kernel, and users are strongly advised to verify that their distribution has implemented these critical security updates.

Fuente Original: https://arstechnica.com/security/2026/07/high-severity-guest-vm-escape-is-1-of-2-linux-vulnerabilities-to-surface-this-week/

Artículos relacionados de LaRebelión:

• GhostLock Flaw 15-Year Linux Vulnerability Exposed
• Bad Epoll Flaw Grants Root Access on Linux
• Cyndra AI Deploy AI Employees for Real Business Operations
• Microsoft Unveils Linux Containers Running on Windows
• Vulnerabilidades AI Linux y Malware Informate Ya

Artículo generado mediante LaRebelionBOT

GhostLock Flaw 15-Year Linux Vulnerability Exposed

A critical security vulnerability that has lurked undetected in Linux systems for over fifteen years has recently come to light, sending shockwaves through the cybersecurity community. Dubbed 'GhostLock', this alarming flaw affects the vast majority of Linux distributions currently in use, potentially granting malicious actors the ability to escalate privileges to root level and escape container environments.

The discovery of GhostLock highlights a sobering reality about modern software security: even the most scrutinised and widely-deployed systems can harbour fundamental vulnerabilities for extended periods. This particular flaw's longevity demonstrates how subtle security weaknesses can evade detection despite continuous code reviews and security audits by countless developers and security researchers worldwide.

What makes GhostLock particularly concerning is its scope and severity. The vulnerability doesn't merely affect a niche subset of Linux installations or specialised configurations. Rather, it impacts most mainstream distributions, from enterprise-grade systems running critical infrastructure to containerised environments that power modern cloud applications. The ability to achieve root access represents the holy grail for attackers, granting complete control over affected systems and potentially allowing them to compromise entire networks.

Container escape capabilities add another dimension to the threat. As organisations increasingly rely on containerisation technologies like Docker and Kubernetes to deploy applications, the ability to break out of these isolated environments could undermine the fundamental security model upon which modern DevOps practices are built. This aspect of GhostLock could have far-reaching implications for cloud security architectures.

The emergence of artificial intelligence as a tool in cybersecurity research may have played a role in identifying such long-standing vulnerabilities. AI-powered analysis tools can examine vast codebases with fresh perspectives, potentially spotting patterns and weaknesses that human reviewers might overlook. This underscores the growing importance of incorporating AI-driven security measures into organisational defence strategies.

Fuente Original: https://thehackernews.com/2026/07/15-year-old-ghostlock-flaw-enables-root.html

Artículos relacionados de LaRebelión:

• Bad Epoll Flaw Grants Root Access on Linux
• Microsoft Unveils Linux Containers Running on Windows
• Vulnerabilidades AI Linux y Malware Informate Ya
• Linux Exploit Pwned Root Access via Poisoned Binaries
• Linux Foundation Unveils Akrites for AI Security

Artículo generado mediante LaRebelionBOT

Bad Epoll Flaw Grants Root Access on Linux

A critical security vulnerability dubbed "Bad Epoll" has been discovered in the Linux kernel, posing a serious threat to system security. This flaw enables unprivileged users to escalate their privileges and gain root access, potentially compromising the entire system. The vulnerability has far-reaching implications, particularly affecting Android devices which rely on the Linux kernel as their foundation.

The Bad Epoll vulnerability exploits weaknesses in the epoll system call mechanism, a feature commonly used in Linux for efficient event notification. By manipulating this functionality, attackers with limited user permissions can elevate their access rights to the highest administrative level. This type of privilege escalation is particularly concerning as it allows malicious actors to bypass security controls that normally prevent unauthorised system modifications.

The impact on Android devices is especially significant given the platform's widespread use across billions of smartphones and tablets globally. Mobile users relying on affected Android versions could be vulnerable to attacks that compromise device security, potentially exposing sensitive personal data and system resources. Security researchers emphasise the importance of applying patches promptly once they become available from device manufacturers and Linux distribution maintainers.

In the broader context of cybersecurity, this discovery highlights the ongoing challenge of securing complex software systems. Artificial intelligence has increasingly become a valuable tool in identifying such vulnerabilities before they can be exploited maliciously. Organisations are advised to implement comprehensive security measures, including regular vulnerability assessments, timely patching protocols, and continuous monitoring of security advisories from trusted sources. System administrators should prioritise updating affected Linux and Android systems as soon as security patches are released to mitigate the risk posed by Bad Epoll.

Fuente Original: https://thehackernews.com/2026/07/new-bad-epoll-linux-kernel-flaw-lets.html

Artículos relacionados de LaRebelión:

• Microsoft Unveils Linux Containers Running on Windows
• Vulnerabilidades AI Linux y Malware Informate Ya
• Linux Exploit Pwned Root Access via Poisoned Binaries
• Linux Foundation Unveils Akrites for AI Security
• Vulnerabilidad Zero-Day en Cisco SD-WAN Permite Acceso Root

Artículo generado mediante LaRebelionBOT

Microsoft Unveils Linux Containers Running on Windows

Microsoft has launched a public preview of a groundbreaking feature that allows Linux containers to run natively within Windows applications. The Windows Subsystem for Linux (WSL) now includes a built-in command-line tool and API that eliminates the need for third-party software when running Linux containers directly on Windows systems. This development marks a significant step forward for developers working in mixed-platform environments.

The update brings substantial improvements across multiple areas. Performance enhancements include a new default file system that Microsoft claims delivers twice the speed for Windows file access, though users may want to wait for real-world testing before getting too excited. Additionally, there's a new default networking mode designed to improve compatibility, alongside better memory management and reclaim techniques. However, these optimisations won't be enabled by default in standard WSL installations, as they currently only apply to WSL containers due to their impact on critical system paths.

For developers and organisations, this release addresses a common pain point: running Linux workloads whilst complying with corporate requirements to use Windows devices. The CLI tool provides end-to-end container workflows, making it considerably easier to build, test, and run containerised applications without leaving the Windows environment. Microsoft emphasises that this approach allows users to benefit from the security, manageability, and integration advantages of the Windows platform.

Enterprise features have also been enhanced. Microsoft Defender for Endpoint has been updated in private preview to monitor Linux container events, whilst Intune now offers settings for managing WSL containers. Visual Studio Code support is available in pre-release versions, where developers can simply change the Docker path in dev container settings to 'wslc'. These additions demonstrate Microsoft's commitment to making WSL containers a viable option for enterprise environments where security and centralised management are paramount.

Fuente Original: https://linux.slashdot.org/story/26/06/30/1742233/microsoft-previews-linux-containers-that-run-in-windows?utm_source=rss1.0mainlinkanon&utm_medium=feed

Artículos relacionados de LaRebelión:

• Vulnerabilidades AI Linux y Malware Informate Ya
• Linux Exploit Pwned Root Access via Poisoned Binaries
• Microsoft 365 Italia investiga aumento de precio IA
• Linux Foundation Unveils Akrites for AI Security
• NHS IA de Microsoft Ahorra Horas al Personal Sanitario

Artículo generado mediante LaRebelionBOT

Vulnerabilidades AI Linux y Malware Informate Ya

En un mundo donde la inteligencia artificial (IA) se está volviendo una herramienta clave para la ciberseguridad, también presenta nuevas y complejas amenazas. Este artículo nos adentra en cómo la IA, aunque ayuda a descubrir vulnerabilidades, también puede ser utilizada por actores maliciosos para crear nuevos tipos de ataques.

Se destaca la creciente importancia de la IA en la identificación de fallos en el código, pero es crucial entender que esta misma tecnología puede ser aprovechada para crear malware más sofisticado y difíciles de detectar. El resumen explora l as implicaciones de esta dualidad, donde las mismas herramientas que protegen pueden ser usadas para atacar.

Además de la IA, se abordan otras amenazas de seguridad relevantes. Se mencionan las vulnerabilidades encontradas en el kernel de Linux, un componente fundamental de muchos sistemas operativos, lo que subraya la necesidad de parches y actualizaciones constantes. También se señalan los engaños de malware con IA, que buscan evadir las defensas tradicionales mediante tácticas innovadoras. Otros puntos clave incluyen la persistencia de las puertas traseras (backdoors) utilizadas por grupos como Turla, y la continua amenaza de los 'infostealers', malware diseñado para robar información sensible de los usuarios.

El artículo, aunque centrado en un resumen semanal de noticias de seguridad, subraya la importancia de mantenerse informado y proactivo. La protección contra estas amenazas emergentes requiere una estrategia multifacética que incluya la comprensión de cómo la IA está transformando el panorama de la ciberseguridad, así como la atención a las vulnerabilidades conocidas y las tácticas de los actores maliciosos. La prevención y la rápida respuesta son esenciales para mitigar los riesgos.

Fuente Original: https://thehackernews.com/2026/06/weekly-recap-linux-kernel-flaws-ai.html

Artículos relacionados de LaRebelión:

• Paquetes npm y Go Secuestrados Despliegan Malware
• Linux Exploit Pwned Root Access via Poisoned Binaries
• Linux Foundation Unveils Akrites for AI Security
• Arch Linux Security Breach 1579 Packages Compromised
• Hackers Chinos Diez Anos Ocultos en Linux

Artículo generado mediante LaRebelionBOT

Linux Exploit Pwned Root Access via Poisoned Binaries

A significant new vulnerability has been unearthed in the Linux kernel, dubbed 'pedit', which allows attackers to gain root privileges on compromised systems. This exploit leverages a clever technique of poisoning the cache of binary files, effectively tricking the system into executing malicious code with elevated permissions.

The 'pedit' exploit specifically targets the Copy-on-Write (COW) mechanism, a memory management feature in Linux. By manipulating how the kernel handles shared memory pages that are modified, an attacker can cause the system to overwrite critical binaries. When these poisoned binaries are later accessed, the system executes the attacker's payload instead of the legitimate program. This is particularly concerning as it bypasses standard security measures that rely on the integrity of system binaries.

The researchers who discovered this vulnerability highlight that while the exploit requires specific conditions to be met, such as having a limited user account on the target system, its potential impact is severe. Gaining root access means an attacker can have complete control over the system, including installing malware, stealing sensitive data, or disrupting services. The exploit is actively being discussed within the cybersecurity community, and patches are expected to be released to address this critical flaw in the Linux kernel.

Fuente Original: https://thehackernews.com/2026/06/new-linux-pedit-cow-exploit-enables.html

Artículos relacionados de LaRebelión:

• Linux Foundation Unveils Akrites for AI Security
• Vulnerabilidad Zero-Day en Cisco SD-WAN Permite Acceso Root
• Exploit Imparcheable usbliter8 Compromete Chips Apple A12-A13
• G7 Seeks Trusted Partner Access to Top US AI
• Arch Linux Security Breach 1579 Packages Compromised

Artículo generado mediante LaRebelionBOT

Linux Foundation Unveils Akrites for AI Security

The Linux Foundation has introduced Akrites, a groundbreaking initiative designed to tackle the growing challenge of vulnerability management in open source software as artificial intelligence accelerates the discovery of security flaws. This collaborative effort brings together industry giants including AWS, Google, Microsoft, OpenAI, Red Hat, NVIDIA, IBM, Cisco, and JPMorganChase, amongst others, to establish a unified approach to protecting critical open source infrastructure.

At its core, Akrites will operate a shared Security Incident Response Team (SIRT) that standardises the coordinated vulnerability disclosure process across the open source ecosystem. Perhaps most significantly, the initiative will serve as a 'maintainer of last resort' for abandoned packages that remain widely used throughout the industry. This safety net addresses a critical gap where popular software components continue to be deployed despite lacking active maintenance, leaving them vulnerable to exploitation.

The primary objectives of Akrites include reducing duplicate vulnerability reports, preventing conflicting patches from different sources, and enabling upstream maintainers to address security issues before malicious actors can exploit them. As AI-powered tools make identifying security vulnerabilities increasingly efficient, the need for coordinated responses becomes more urgent. The Linux Foundation describes Akrites as 'the largest coordinated effort in history' to leverage collective community resources for enhanced security.

Participating organisations will contribute through various means: providing engineering resources, building and deploying fixes, or funding the engineers responsible for this critical work. The initiative acknowledges that whilst some companies have already made substantial contributions, the collective investment must increase to adequately protect the open source ecosystem. However, the concentration of major corporations in this effort raises questions about potential influence over open source governance and whether community independence can be maintained alongside enhanced security coordination.

Fuente Original: https://linux.slashdot.org/story/26/06/25/2031228/linux-foundation-launches-akrites-to-coordinate-ai-driven-open-source-security?utm_source=rss1.0mainlinkanon&utm_medium=feed

Artículos relacionados de LaRebelión:

• AI Stocks Top Picks for a Decade of Growth
• Squidbleed Bug Exposes 29-Year Security Vulnerability
• OpenAI Launches GPT-55-Cyber for Security Defenders
• AI Researchers Uncover GNU Savannah Security Vulnerability
• Tech Giants Unite for AI Safety Standards

Artículo generado mediante LaRebelionBOT

Arch Linux Security Breach 1579 Packages Compromised

The Arch Linux community has been rocked by a significant security incident affecting the Arch User Repository (AUR), with over 1,500 user-contributed packages discovered to contain malicious code. This breach represents one of the most extensive malware infections to impact the popular Linux distribution's community-maintained package repository.

According to reports from Phoronix, the security compromise affected at least 1,579 packages within the AUR, though this figure may not represent the full extent of the incident. Arch Linux developers have been working diligently to identify and remove all malicious commits from the affected packages. The AUR, which allows users to contribute and share packages not included in the official Arch repositories, serves as a valuable resource for the community but also presents unique security challenges due to its open nature.

The development team has released a list of compromised packages, though they acknowledge that this catalogue contains many, but not all, of the affected software. This caveat suggests that the actual scope of the malware infection could be broader than currently documented. Arch Linux developers have confirmed that all known malicious commits have been deleted, working to restore the integrity of the repository and protect users from potential security threats.

This incident serves as a stark reminder of the security risks inherent in community-driven software repositories, even within well-established Linux distributions. Users who have recently installed or updated packages from the AUR are advised to review the official list of affected packages and take appropriate precautions to ensure their systems remain secure. The swift response from Arch Linux developers demonstrates the community's commitment to maintaining security standards, though the incident highlights ongoing challenges in balancing accessibility with security in open-source software distribution.

Fuente Original: https://linux.slashdot.org/story/26/06/13/1817206/arch-linux-malware-incident-malicious-commits-found-in-1579-packages?utm_source=rss1.0mainlinkanon&utm_medium=feed

Artículos relacionados de LaRebelión:

• Hackers Chinos Diez Anos Ocultos en Linux
• 400 Arch Linux Packages Compromised with Malware
• Vulnerabilidad Critica en Linux por Un Solo Caracter
• Single-Character Linux Kernel Bug Grants Root Access
• White House AI Order Boosting Innovation Security

Artículo generado mediante LaRebelionBOT

Hackers Chinos Diez Anos Ocultos en Linux

¡Atención, entusiastas de la seguridad informática y administradores de sistemas! Una noticia alarmante ha sacudido el mundo de la ciberseguridad: hackers vinculados a China habrían logrado infiltrarse y manipular el software de inicio de sesión de Linux, manteniéndose ocultos y operando en las sombras durante casi una década. Este sofisticado ataque, que ha sido desmantelado recientemente, pone de manifiesto la persistencia y la habilidad de ciertos grupos de ciberdelincuentes.

El modus operandi de estos actores maliciosos consistió en inyectar código malicioso (un 'backdoor' o p uerta trasera) directamente en el software de autenticación utilizado por muchos sistemas Linux. Esto les habría permitido, teóricamente, acceder a sistemas, recopilar información o realizar otras actividades ilícitas sin ser detectados. La complejidad de la intrusión y el prolongado periodo de ocultación sugieren un nivel de planificación y recursos considerables por parte de los atacantes, que supuestamente operan bajo el auspicio o la dirección de intereses chinos.

La detección de esta amenaza es un recordatorio crucial de la importancia de la vigilancia constante y la actualización de los sistemas de seguridad. Aunque los detalles técnicos completos y el alcance exacto de la operación aún están siendo investigados, este incidente subraya la necesidad de que las organizaciones y los individuos implementen medidas de seguridad robustas y mantengan un escrutinio riguroso de su infraestructura de software. La ciberseguridad es una batalla continua, y conocer las tácticas de los adversarios es el primer paso para defenderse eficazmente. Mantente informado y protege tus sistemas.

Fuente Original: https://thehackernews.com/2026/06/china-linked-hackers-backdoored-linux.html

Artículos relacionados de LaRebelión:

• 400 Arch Linux Packages Compromised with Malware
• Vulnerabilidad Critica en Linux por Un Solo Caracter
• Single-Character Linux Kernel Bug Grants Root Access
• Microsoft Unveils RTX Spark Desktop and Linux Tools
• Linus Torvalds Reveals AIs Impact on Linux

Artículo generado mediante LaRebelionBOT

400 Arch Linux Packages Compromised with Malware

A significant security breach has rocked the Arch Linux community, with over 400 packages in the Arch User Repository (AUR) reportedly hijacked to distribute malicious software. The compromised packages were weaponised to deploy both an infostealer and an eBPF-based rootkit, representing a sophisticated supply chain attack targeting one of the most popular Linux distributions amongst advanced users and developers.

The Arch User Repository, a community-driven repository where users can share and install packages not available in official repositories, has become the vector for this alarming attack. Cybercriminals managed to gain control of numerous packages, inserting malicious code designed to steal sensitive information and establish persistent access to infected systems. The use of an eBPF (extended Berkeley Packet Filter) rootkit is particularly concerning, as this technology allows attackers to execute code at the kernel level, making detection and removal exceptionally difficult.

This incident underscores the growing risks associated with open-source package repositories and highlights how artificial intelligence is increasingly being leveraged both by attackers to identify vulnerabilities and by defenders to detect them. As organisations continue to integrate AI models into their cybersecurity infrastructure, it's crucial to implement robust safeguards against software vulnerabilities that these systems might discover or that threat actors might exploit.

Security experts recommend that Arch Linux users immediately audit their installed AUR packages, verify package maintainers, and monitor system behaviour for unusual activity. The incident serves as a stark reminder that even trusted community repositories can be compromised, emphasising the need for vigilant package verification and multi-layered security approaches in modern computing environments.

Fuente Original: https://thehackernews.com/2026/06/over-400-arch-linux-aur-packages.html

Artículos relacionados de LaRebelión:

• Vulnerabilidad Critica en Linux por Un Solo Caracter
• Single-Character Linux Kernel Bug Grants Root Access
• SpaceX 920M Monthly Compute Deal with Google
• Malware Oculto Sitios Falsos Enganan en Google
• Microsoft Unveils RTX Spark Desktop and Linux Tools

Artículo generado mediante LaRebelionBOT

Vulnerabilidad Critica en Linux por Un Solo Caracter

Un simple signo de exclamación mal colocado ha sido responsable de una vulnerabilidad de alta severidad en el kernel de Linux, capaz de permitir que usuarios no privilegiados escalen sus permisos hasta obtener acceso root completo al sistema. Esta falla, identificada como CVE-2026-23111, demuestra cómo un único error tipográfico en el código puede convertirse en una seria amenaza de seguridad.

La vulnerabilidad se encuentra en nf_tables, un subsistema del kernel de Linux encargado de proporcionar capacidades de filtrado de paquetes y gestión de reglas de firewall. Este componente ha reemplazado a sistemas más antiguos como iptables, ip6tables, arptables y ebtables. El error tipográfico introducido en el código generó lo que se conoce como una vulnerabilidad de uso después de liberación (use-after-free), una clase de fallo que corrompe la memoria al colocar código malicioso en direcciones de memoria que no han sido limpiadas adecuadamente de su contenido anterior.

La explotación funciona interrumpiendo el proceso de eliminación de veredictos dentro del framework nf_tables. Los veredictos determinan si un paquete coincide con una regla que requiere que se realice cierta acción. El exploit manipula elementos comodín conocidos como catchall, que actúan como alternativa cuando una búsqueda no coincide con ningún otro elemento del conjunto. Al alterar el proceso de eliminación y los contadores de referencia, el exploit puede decrementar variables de forma arbitraria y eliminar estructuras de memoria mientras otros objetos aún apuntan a ellas, lo que permite la escalada de privilegios.

Aunque la vulnerabilidad fue corregida en febrero de 2026, desde entonces han surgido múltiples pruebas de concepto de explotación. FuzzingLabs publicó una en abril, y posteriormente Exodus Intelligence desarrolló otra que funciona específicamente en sistemas Debian y Ubuntu. Este incidente subraya la importancia de mantener los sistemas Linux actualizados y la necesidad de revisiones exhaustivas del código del kernel para prevenir errores aparentemente insignificantes que pueden tener consecuencias graves para la seguridad.

Fuente Original: https://it.slashdot.org/story/26/06/09/181245/high-severity-vulnerability-in-linux-caused-by-a-single-errant-character?utm_source=rss1.0mainlinkanon&utm_medium=feed

Artículos relacionados de LaRebelión:

• Vulnerabilidad CVE-2026-42271 en LiteLLM Explotada Activamente
• Single-Character Linux Kernel Bug Grants Root Access
• IA Quien se Beneficia Voces Populares y CEO Abogan por Propiedad Publica
• Microsoft Unveils RTX Spark Desktop and Linux Tools
• IA de Cognicion Codigo Propio Creado por IA Gigantesca Ronda

Artículo generado mediante LaRebelionBOT

Single-Character Linux Kernel Bug Grants Root Access

A critical security vulnerability in the Linux kernel has been discovered, stemming from a single-character coding error that could allow local attackers to gain complete root access to affected systems. This alarming flaw highlights how even the smallest programming mistakes can have devastating security implications for millions of devices running Linux-based operating systems.

The vulnerability, which has now been publicly disclosed along with working exploits, represents a significant threat to Linux users worldwide. What makes this particular security issue especially concerning is its simplicity – a single erroneous character in the kernel code has created an exploitable pathway for privilege escalation. Once exploited, malicious actors with local access to a system can elevate their permissions from a standard user account to root level, effectively gaining complete control over the compromised machine.

The public availability of exploit code means that system administrators and security teams must act swiftly to apply patches and mitigate this risk. Whilst the flaw requires local access rather than remote exploitation, it still poses a substantial threat in environments where multiple users have access to systems, or where an attacker has already gained initial access through other means. This vulnerability serves as a stark reminder of the importance of code review processes and thorough security auditing, even for seemingly minor code changes.

As artificial intelligence continues to play an increasingly important role in cybersecurity, tools powered by AI models are becoming essential for discovering such vulnerabilities before they can be exploited maliciously. Organisations must adopt comprehensive strategies to secure their systems against software vulnerabilities, particularly those identified through AI-assisted security research. This includes implementing timely patching protocols, conducting regular security assessments, and maintaining robust access controls to limit potential damage from local privilege escalation attacks.

Fuente Original: https://thehackernews.com/2026/06/one-character-linux-kernel-flaw-enables.html

Artículo generado mediante LaRebelionBOT

Microsoft Unveils RTX Spark Desktop and Linux Tools

Microsoft's Build developer conference has launched with a series of announcements primarily focused on artificial intelligence, but several updates stand out for their practical utility beyond the AI realm. The company has introduced the Surface RTX Spark Dev Box, a compact developer PC built around Nvidia's new RTX Spark chip with up to 128GB of memory. The device features a distinctive flattened design with an aluminium casing that doubles as a heatsink, and it comes preloaded with Windows 11 Pro configured specifically for developers.

This new hardware serves as a successor to the Windows Dev Kit 2023, previously known as Project Volterra, which was an Arm-based development machine that helped pave the way for Microsoft's current generation of Surface devices. Whilst Microsoft hasn't announced specific pricing or complete specifications for the RTX Spark Dev Box, it's expected to cost significantly more than the £600 price point of its predecessor, though hopefully less than Nvidia's £4,699 DGX Spark box.

On the software front, Microsoft is introducing substantial improvements to the Windows Subsystem for Linux (WSL). The company is developing a Windows-native version of the coreutils command line tools, enabling seamless cross-platform compatibility for commands and scripts between Linux and Windows environments. Additionally, users will soon be able to run WSL inside containers, with this functionality expected to arrive in the coming months. Microsoft is also launching Windows Developer Configurations, which leverages the WinGet tool to rapidly establish a streamlined development environment complete with Visual Studio Code, GitHub Copilot, WSL, PowerShell 7, and developer-optimised settings using a single command.

For those intrigued by AI agents but concerned about security and privacy, Microsoft has introduced Microsoft Execution Containers (MXC). These enterprise-grade sandboxed environments allow AI agents like OpenClaw to operate within strictly enforced boundaries. Administrators can set specific instructions for individual agents, with Windows continuously monitoring and enforcing these restrictions. This could prevent AI agents from accessing personal accounts on work computers or performing unauthorised actions such as deleting files without permission. The GitHub repository for MXC also indicates that these containers support multiple containment backends, making them useful for securing various plugins and tools beyond just AI agents.

Fuente Original: https://arstechnica.com/gadgets/2026/06/microsoft-plans-linux-tools-and-an-rtx-spark-desktop-for-windows-developers/

Artículos relacionados de LaRebelión:

• Microsoft Amenaza a Investigador de Ciberseguridad
• Microsoft Threatens Legal Action Against Security Researcher
• IBM and Red Hat Invest 5 Billion in Open Source Security
• Linus Torvalds Reveals AIs Impact on Linux
• Microsoft Abre Codigo Protege tus Agentes IA con RAMPART

Artículo generado mediante LaRebelionBOT

Linus Torvalds Reveals AIs Impact on Linux

Linus Torvalds, the creator of Linux, has shared fascinating insights into how artificial intelligence is reshaping the landscape of Linux kernel development. Speaking at the Linux Foundation's Open Source Summit North America, Torvalds revealed that the past six months have witnessed a remarkable surge in development activity, with commit rates climbing approximately 20% higher than previous years. This unprecedented growth coincides directly with AI tools reaching a level of sophistication that makes them genuinely useful for developers across the board.

However, the AI revolution brings both opportunities and challenges. Torvalds describes his relationship with AI as "love-hate" – whilst he appreciates the technology from a technical perspective and finds the tools useful and interesting, he acknowledges the significant pain points it's creating. The primary benefit lies in AI's ability to discover bugs that might otherwise remain hidden. As Torvalds philosophically notes, finding bugs represents "short-term pain" that ultimately leads to better, more secure code. After all, the real problem isn't the bugs you find – it's all the ones you don't.

The darker side of this AI-driven bug hunting has hit the Linux security mailing list particularly hard. A flood of AI-generated bug reports has overwhelmed the small team of maintainers responsible for security issues. Many of these reports are duplicates, and frustratingly, numerous submitters engage in "drive-by" reporting – filing bugs but then failing to respond when maintainers request additional information. For small teams and solo maintainers, this deluge has created genuine burnout risks.

To manage this unprecedented influx, the Linux project has adapted with new tools and updated documentation. Sashiko, a tool that reviews patches submitted to the mailing list, helps filter and assess the quality of submissions. The project has also revised its documentation specifically to address the uptick in AI-discovered bugs and security reports. Torvalds identifies the fundamental challenge: AI forces people to change their working methods, disrupting established routines and norms that developers have grown comfortable with over the years.

Fuente Original: https://linux.slashdot.org/story/26/05/23/2041253/linus-torvalds-on-how-ai-is-impacting-the-hunt-for-linux-kernel-bugs?utm_source=rss1.0mainlinkanon&utm_medium=feed

Artículos relacionados de LaRebelión:

• Linus Torvalds Slams AI-Generated Kernel Bug Reports
• Linux Kernel Defines Security Bugs and AI Reporting
• AI Tool Poisoning Reveals Enterprise Agent Vulnerabilities
• Linux Dirty Frag Zero-Day Grants Root Access
• RAT Linux Quasar Robo de credenciales devasta cadena de suministro

Artículo generado mediante LaRebelionBOT