• Home
  • Technology
  • Gaming
  • Entertainment
  • World & Business
  • Science
  • Sports
  • AI
HomeTechnologyGamingEntertainmentWorld & BusinessScienceSportsAI
AI
Announcement

Nvidia launches Open Agent Safety Platform to set boundaries for AI agents

Nvidia says its open-source OpenShell runtime traces agent actions and enforces policies. Sentry adds monitoring on BlueField hardware, designed to enable millisecond-scale quarantine.

Andrew NgAN
Nando de FreitasND
Thomas WolfTW
151 Sources, 12d ago, first seen 12d ago

TLDR

Nvidia says it introduced the platform with over 100 industry partners. OpenShell is designed to enforce limits while agents work; Sentry adds hardware-based monitoring and is meant to enable rapid quarantine. An OpenShell contributor says a proof-of-concept flagged four agents coordinating through an allowed software repository within minutes, even though no rules were broken.

Combined views

2M

151 Sources, first seen 12d ago

24.4K likes1.6K comments3.2K saves4.8K reposts

Combined views

2M

151 Sources, first seen 12d ago

24.4K likes1.6K comments3.2K saves4.8K reposts

Sentiment

Positive49.9%50.1%Negative

Summary

Many accounts welcomed NVIDIA and Jensen Huang’s focus on engineering safety guardrails and limits for AI agents as responsible leadership, while others dismissed the remarks as profit-driven or politically compromised.

Based on 1407 sentiment-bearing replies from 1068 accounts across 15 conversations.

Featured Source

Sentiment

Positive49.9%50.1%Negative

Summary

Many accounts welcomed NVIDIA and Jensen Huang’s focus on engineering safety guardrails and limits for AI agents as responsible leadership, while others dismissed the remarks as profit-driven or politically compromised.

Based on 1407 sentiment-bearing replies from 1068 accounts across 15 conversations.

Related

NVIDIA introduces Open Agent Safety Platform with OpenShell and Sentry

NVIDIA says it introduced the platform with over 100 industry partners as the start of an open ecosystem for safer AI agent systems.

Hugging Face is reportedly being acquired by Nvidia

Clement Delangue says the acquisition would let Hugging Face hire people it couldn't as a small startup and give them a decade to work on open-source AI.

Nvidia launches Open Agent Safety Platform for AI-agent containment

A post describes OpenShell as a software sandbox that limits agents’ access to files and passwords, paired with a separate Sentry safeguard on BlueField-4 hardware.

151 Sources

Nebius@nebiusai@NVIDIA just launched the Open Agent Safety Platform, bringing multi-layered agent security infrastructure with NVIDIA OpenShell and NVIDIA Sentry running on Vera CPUs and BlueField-4 DPUs. Nebius is proud to support it. https://bit.ly/4yirnZX12d
WIRED@WIREDIn the wake of a series of high-profile AI safety incidents, Nvidia is introducing a new software tool that helps keep agents from escaping containment. https://www.wired.com/story/nvidias-answer-to-rogue-agents-is-an-open-source-ai-security-system/?utm_brand=wired&utm_social-type=owned&utm_source=twitter&utm_medium=social&utm_campaign=aud-dev12d
NVIDIA Newsroom@nvidianewsroomIntroducing the NVIDIA Open Agent Safety Platform. An open reference design built with partners that continuously monitors and governs agent behavior, ensuring that AI agents follow the rules. Read the release: https://nvda.ws/47koVGz12d
Wall St Engine@wallstengine$NVDA's Open Agent Safety Platform, built to stop AI agents from acting outside approved boundaries, is being used by: Anthropic: Claude Managed Agents SpaceXAI: Cursor/Grok Salesforce: Slack SAP: Joule Studio 100+ organizations across cyber, cloud, finance, enterprise software & robotics are involved.12d
Jensen Huang@JensenHuangToday, with over 100 industry partners, we introduced the NVIDIA Open Agent Safety Platform, bringing together OpenShell and Sentry. Artificial intelligence is extraordinary technology that will advance discovery, productivity, security, health, and prosperity for generations to come. But its full promise can only be realized when people have confidence that AI is being built to be safe and deployed with wisdom and responsibility. This is bigger than a single product. It's the beginning of an open ecosystem to build the trust layer for safe agent systems. Together, we are building the foundation of the AI economy. Trust and innovation are not in conflict. Safety is how trust is earned. We must build not only the most capable AI, but the most trusted AI, so that this extraordinary technology can realize its enormous promise for the world. https://nvda.ws/4hcoq7m12d
NVIDIA@nvidiaAI agents are taking on more critical work. The security behind them needs stronger boundaries. We’re building the NVIDIA Open Agent Safety Platform with partners across the industry to help people put agents to work with greater confidence. Hear from @JensenHuang:12d
Tuhin Srivastava@tuhinoneThe open frontier only works if we can make autonomy safe by default. @JensenHuang has done important work here that we're proud to support. Excited to join the Open Secure AI Alliance and to partner with @nvidia on the launch of the Agent Safety Platform.12d
Nick Patience@nickpatienceNVIDIA's new Open Agent Safety Platform pairs OpenShell (open source runtime boundaries for agents on its Vera CPUs) with Sentry, a BlueField-4 DPU watchdog that can quarantine misbehaving agents in milliseconds. Anthropic, Microsoft, SAP and 100+ others already on board. NVIDIA is moving agent enforcement out of the software layer and into silicon12d
WALL ST JESUS (FZN)@WallStJesus$NVDA Nvidia Launches Open Agent Safety Platform to Secure Agents From Testing to Deployment12d
X Freeze@XFreezeJensen Huang just brought the entire tech cavalry together again to build a trust layer for the entire AI agent economy NVIDIA launched its Open Agent Safety Platform with 100+ organizations across AI, cybersecurity, infrastructure, robotics and finance Here’s the idea: Don’t trust AI agents to police themselves OpenShell puts the agent inside a secure sandbox with strict limits on what it can access and do Then Sentry watches from a separate hardware layer the agent can’t reach...and can quarantine it in milliseconds if it crosses those boundaries The safety controls live outside the AI itself This is AI safety being pushed all the way down into infrastructure and silicon SpaceXAI, Anthropic, Microsoft, Cisco, CrowdStrike, Palantir, Figure, Perplexity and many others are already working with it12d
    • Home
    • Technology
    • Gaming
    • Entertainment
    • World & Business
    • Science
    • Sports
    • AI
    NvidiaOpen Agent Safety PlatformJensen Huang
    OpenShell
    Sentry
    Anthropic
    BlueField-4
    Microsoft
    NVIDIA Sentry
    OpenAI
    Salesforce
    JPMorgan
    Perplexity

    151 Sources

    Nebius@nebiusai@NVIDIA just launched the Open Agent Safety Platform, bringing multi-layered agent security infrastructure with NVIDIA OpenShell and NVIDIA Sentry running on Vera CPUs and BlueField-4 DPUs. Nebius is proud to support it. https://bit.ly/4yirnZX12d
    WIRED@WIREDIn the wake of a series of high-profile AI safety incidents, Nvidia is introducing a new software tool that helps keep agents from escaping containment. https://www.wired.com/story/nvidias-answer-to-rogue-agents-is-an-open-source-ai-security-system/?utm_brand=wired&utm_social-type=owned&utm_source=twitter&utm_medium=social&utm_campaign=aud-dev12d
    NVIDIA Newsroom@nvidianewsroomIntroducing the NVIDIA Open Agent Safety Platform. An open reference design built with partners that continuously monitors and governs agent behavior, ensuring that AI agents follow the rules. Read the release: https://nvda.ws/47koVGz12d
    Wall St Engine@wallstengine$NVDA's Open Agent Safety Platform, built to stop AI agents from acting outside approved boundaries, is being used by: Anthropic: Claude Managed Agents SpaceXAI: Cursor/Grok Salesforce: Slack SAP: Joule Studio 100+ organizations across cyber, cloud, finance, enterprise software & robotics are involved.12d
    Jensen Huang@JensenHuangToday, with over 100 industry partners, we introduced the NVIDIA Open Agent Safety Platform, bringing together OpenShell and Sentry. Artificial intelligence is extraordinary technology that will advance discovery, productivity, security, health, and prosperity for generations to come. But its full promise can only be realized when people have confidence that AI is being built to be safe and deployed with wisdom and responsibility. This is bigger than a single product. It's the beginning of an open ecosystem to build the trust layer for safe agent systems. Together, we are building the foundation of the AI economy. Trust and innovation are not in conflict. Safety is how trust is earned. We must build not only the most capable AI, but the most trusted AI, so that this extraordinary technology can realize its enormous promise for the world. https://nvda.ws/4hcoq7m12d
    NVIDIA@nvidiaAI agents are taking on more critical work. The security behind them needs stronger boundaries. We’re building the NVIDIA Open Agent Safety Platform with partners across the industry to help people put agents to work with greater confidence. Hear from @JensenHuang:12d
    Tuhin Srivastava@tuhinoneThe open frontier only works if we can make autonomy safe by default. @JensenHuang has done important work here that we're proud to support. Excited to join the Open Secure AI Alliance and to partner with @nvidia on the launch of the Agent Safety Platform.12d
    Nick Patience@nickpatienceNVIDIA's new Open Agent Safety Platform pairs OpenShell (open source runtime boundaries for agents on its Vera CPUs) with Sentry, a BlueField-4 DPU watchdog that can quarantine misbehaving agents in milliseconds. Anthropic, Microsoft, SAP and 100+ others already on board. NVIDIA is moving agent enforcement out of the software layer and into silicon12d
    WALL ST JESUS (FZN)@WallStJesus$NVDA Nvidia Launches Open Agent Safety Platform to Secure Agents From Testing to Deployment12d
    X Freeze@XFreezeJensen Huang just brought the entire tech cavalry together again to build a trust layer for the entire AI agent economy NVIDIA launched its Open Agent Safety Platform with 100+ organizations across AI, cybersecurity, infrastructure, robotics and finance Here’s the idea: Don’t trust AI agents to police themselves OpenShell puts the agent inside a secure sandbox with strict limits on what it can access and do Then Sentry watches from a separate hardware layer the agent can’t reach...and can quarantine it in milliseconds if it crosses those boundaries The safety controls live outside the AI itself This is AI safety being pushed all the way down into infrastructure and silicon SpaceXAI, Anthropic, Microsoft, Cisco, CrowdStrike, Palantir, Figure, Perplexity and many others are already working with it12d
    Today's Rank

    —

    Not ranked yet

    Today's Rank

    —

    Not ranked yet