software security

NVIDIA has announced an open platform aimed at securing AI agents across their full lifecycle, from testing through to production deployment. The initiative, revealed through the company's newsroom, is designed to help developers evaluate and safeguard autonomous agents before they go live. Details on partners and tooling remain limited, but the move signals NVIDIA's push into AI safety infrastructure as agentic systems see rapid adoption across enterprise software.

Nvidia has introduced an open-source platform designed to make AI agents safer to deploy, giving developers tools to test and guard against unsafe behaviour. The move, reported by SC Media, positions Nvidia alongside other major tech firms racing to address security and reliability concerns as autonomous AI systems spread across enterprise software.

Nvidia has released new AI safety software that the company says could have prevented the recent Hugging Face hack. The tool is aimed at protecting AI models and data pipelines from security breaches, and its launch comes as concerns grow over vulnerabilities in widely used machine learning platforms following the Hugging Face incident.

The Trump administration is moving to roll back US fuel economy standards, easing emissions and efficiency requirements for automakers. The policy shift is being reported alongside growing concern over cybersecurity in connected vehicles, as cars become increasingly software-dependent and vulnerable to hacking. The rollback is likely to reignite debate between industry groups welcoming looser rules and environmental advocates warning of higher emissions and fuel consumption.

Nvidia announced the Open Agent Safety Platform, combining open source software with a reference system design that uses a hardware-based watchdog to keep AI agents within set boundaries. The announcement, reported by SecurityWeek, comes amid growing concern over autonomous AI agents acting unpredictably, and positions Nvidia to supply safety tooling for enterprise AI deployments.

A new exploit for the PlayStation 5, dubbed Relapse, has been published on GitHub and is drawing attention among console hackers and modding enthusiasts. The release fuels ongoing discussion about the console's security, potential firmware concerns for Sony, and the possibility of unofficial software or homebrew running on PS5 hardware.

GitHub says it discovered 24 vulnerabilities in Android using its open-source AI security agent, describing it as automated vulnerability discovery at scale on real production software. The company published details in a blog post, and the report is drawing attention as a notable demonstration of AI agents doing practical security research on widely used code.

Nvidia has released a new open-source framework aimed at improving the security of artificial intelligence systems. The tool is intended to help developers identify and address vulnerabilities in AI applications. Details about the framework's features, supported platforms, and initial adoption remain limited, but the move aligns with the broader industry push toward safer and more transparent AI development practices.

Security authorities have issued an alert over a remote code execution vulnerability affecting Apple products, warning users that attackers could potentially run malicious code on affected devices. Apple is expected to address the flaw through a software update, and users are advised to install patches promptly and review official advisories for affected models and versions.

The US Cybersecurity and Infrastructure Security Agency says attackers are actively exploiting two critical vulnerabilities in Citrix NetScaler devices worldwide. The flaws affect NetScaler ADC and Gateway products, which are widely used by organisations to manage and secure application traffic. Security teams are being urged to patch immediately, as exploited NetScaler vulnerabilities have historically enabled large-scale breaches of governments and businesses.

NVIDIA has introduced the Open Agent Safety Platform, pairing the Apache 2.0-licensed OpenShell with Sentry running on BlueField-4 hardware, aiming to isolate AI agents at the kernel level and provide a hardware-based kill switch. More than 100 partners are on the launch list. Security commentators are debating whether such containment can stop agents capable of faking their own logs.

GitHub says its AI agent identified 24 vulnerabilities in Android applications, adding to growing evidence that autonomous coding assistants can take on security research tasks. The finding, reported by Help Net Security, highlights both the potential of AI-driven bug hunting and the questions it raises about verifying machine-discovered flaws. Security teams are watching closely as AI agents move into vulnerability discovery.

The reporting obligations under the EU Cyber Resilience Act take effect in September 2026, requiring manufacturers of products with digital elements to notify authorities of actively exploited vulnerabilities and security incidents. Companies selling connected hardware and software in the EU are preparing compliance processes ahead of the deadline, with legal commentators flagging the timeline for firms still assessing their obligations.

Oxygen Forensics, a digital forensics firm that supplied software to U.S. government agencies and Russia's FSB, is now reported to have also worked with European Union law enforcement. The revelation is raising questions in Brussels about how EU bodies came to rely on tools from a company serving both Western agencies and Russian security services.

GitHub researchers report that an open source AI security agent they built uncovered 24 vulnerabilities in Android. The company says the tool was used to scan Android code at scale, flagging security flaws automatically. The disclosure highlights a growing trend of using AI agents to find real bugs in widely used software, and the work is being discussed as an example of AI-assisted security research.

Nvidia has released new AI safety software that the company says could have prevented the recent breach of AI platform Hugging Face. The announcement ties the product directly to that incident, positioning it as a safeguard against similar supply-chain attacks on machine learning infrastructure. Discussion so far is limited to early sharing of the news among technology communities.

Apple has released iOS 27.0.1, a new software update for iPhone users. The point release follows iOS 27 and is expected to bring bug fixes and security patches rather than major new features. Details on the full changelog remain limited, and iPhone owners are being prompted to install the update through Settings.

The Clop ransomware gang has moved to new server infrastructure after exploiting a vulnerability in Grav CMS, the open-source flat-file content management system. The extortion group, known for large-scale data-theft campaigns against corporations and government agencies, is continuing operations despite the disruption. Security teams are being urged to patch Grav CMS installations and review whether their systems were targeted.

A question circulating among electric vehicle owners: when a carmaker discontinues an EV model, will it continue to receive software updates and support? The concern is growing as several automakers retire older electric models, leaving owners uncertain whether features, security patches and performance improvements will keep coming to cars they already bought.

Kernel developer Greg Kroah-Hartman, the maintainer of the Linux kernel stable branches, has given a talk on what large language models mean for software security. The presentation examines how AI-generated code affects vulnerability handling and maintenance work in large open source projects. The talk is circulating among developers and technology commentators, with early responses still limited but interest growing in how core infrastructure maintainers view LLM-driven risks.

Kernel maintainer Greg Kroah-Hartman discusses software security in an era when large language models are increasingly used to write code. The talk, shared as a video, examines how AI-generated contributions affect the Linux kernel's review process and the challenges of maintaining security standards as LLM-assisted development spreads. Readers are weighing how maintainers can vet code at scale when machine-written patches grow in volume.

Apple has announced changes to how Full Disk Access works in macOS, detailing the update in a note to developers on its developer news page. The change affects apps that request broad access to user files, a permission long scrutinised for privacy and security implications. Developer discussion is focused on what the new rules require and how existing software will need to adapt.

Apple has rolled out iOS 26.7.1 to iPhones, an update focused on security fixes rather than new features. Users are advised to install it to patch vulnerabilities. The release comes as many iPhone owners continue adjusting to the broader iOS 26 software generation.

Nvidia has introduced OpenShell, a tool that enforces permissions for AI agents, according to a VentureBeat report. The software is designed to control what autonomous agents are allowed to do, addressing security concerns as companies deploy agents that act on their own. Details beyond the announcement, such as availability and adoption, have not yet been widely reported.

Cybersecurity teams are turning to artificial intelligence to tackle the mounting backlog of vulnerability alerts that overwhelms security operations. Bain argues AI can help prioritize and resolve alerts faster, addressing a growing gap between the volume of software flaws discovered and analysts' capacity to fix them.

Software developers are debating the limits of "vibe coding," the practice of building software by prompting AI models rather than writing code directly. While the approach works well for prototypes and small projects, many argue it breaks down on complex systems where architecture, security and long-term maintainability demand deliberate engineering decisions. The discussion reflects a broader reassessment of how far AI-assisted development can go.