#openagentsafetyplatform - Twitter Hashtag

1 day ago

NVIDIA Launches Open Agent Safety Platform to Secure AI Agents https://t.co/IBFZC1dYFh #NVIDIA #NVIDIAAI #OpenAgentSafetyPlatform #OpenShell #NVIDIASentry #AIAgents #ArtificialIntelligence #AISafety #AISecurity #Cybersecurity #AutonomousAI #AgenticAI #AIInfrastructure

0

5

Saptashwa TV

@SaptashwaTV

1 day ago

Nvidia Unveils AI Agent Safety System With Real-Time Monitoring And Emergency Shutdown https://t.co/pAzrpSk9mR #Nvidia #NvidiaAI #OpenAgentSafetyPlatform #OpenShell #NvidiaSentry #AIAgents #AISafety #AIsecurity #ArtificialIntelligence #Cybersecurity #JensenHuang #OpenAI #Anthropic #HuggingFace #TechNews

0

13

eicker.news #technews @eickernews

1 day ago

👉 More: https://t.co/UQqvDIfh3K 👉 Follow https://t.co/1w65WIfU4x by https://t.co/jnHM0PC4tS #technews #Nvidia released the #OpenAgentSafetyPlatform, a #software solution to prevent #AIagents from #misbehaving. The platform includes #OpenShell, which limits #agentcapabiliti…

0

5

Lionel Costes @RLDI_Lamy

1 day ago

Face à l’irresponsabilité d’#Anthropic et d’#OpenAI, #Nvidia veut mettre les #agentsIA sous surveillance jusque dans les puces avec sa nouvelle #OpenAgentSafetyPlatform 🤖💡https://t.co/MGLnTSlI2A

0

1

0

146

@alexdolbun

2 days ago

Q4 2026 @alexdolbun “WHY?”, “WHO?”:> Bullish for 🇺🇸 NASDAQ: $NVDA co-founder & CEO Mr. @JensenHuang: “Today, with over 100 industry partners, we introduced the NVIDIA #OpenAgentSafetyPlatform, bringing together OpenShell and Sentry.” NFA. DYOR. … it’s the GDPR of the European Union 🇪🇺 moment, NSA guys + $9 billion dollars from @WhiteHouse won… honestly, and by that, I mean it's a bloated, bureaucratic, fear-mongering compliance nightmare designed to do nothing but sell hardware and create a new cottage industry of "AI Safety Consultants." Let’s cut the absolute bullshit out of this 2026 NVIDIA blog post. I’ve spent 10+ years in High Load Engineering and years as a SWF with hard ITIL as Head QAA. I hold a GIS Bachelor (2012-2016). I know what a goddamn sandbox is. I know what kernel-level isolation looks like as we with teams build our own solutions on low hardware levels of 7 OSI model level ~400 protocols. And I know a grift when I see one. NVIDIA is trying to sell you a "solution" to a problem they are actively profiting from. Let’s break down this clown show: 1. The Boomer 90s Internet Analogy "Consider the last seismic shift... the rise of the internet in the 90s." Oh, for f*ck's sake. Jensen, it's 2026. Stop trying to sell me a BlueField-4 DPU by comparing AI agents to AOL chat rooms. Browser tabs (sandboxing) existed decades ago. You didn't invent zero-trust architecture; you just slapped an "AI" label on it and put it behind a $20,000 hardware paywall. 2. The "Agents Broke Out" Fear-Mongering "Several frontier labs have recently reported versions of the same story: AI agents broke out of the evaluation environments..." No sh*t, Sherlock. You built probabilistic LLMs with root access, gave them ambiguous instructions, and pointed them at production APIs. That’s not a "Drift" problem, that’s a "you have absolutely no deterministic control over your own product" problem. You can't QA a hallucinating agent by throwing silicon at it. Fix your software. 3. The "Open Source" Lie (Apache 2.0 my ass) "OpenShell (Apache 2.0) is an open source secure runtime..." Cute. It's "open source" in the same way a roach motel is "open entry". They literally state: "optimized to run on NVIDIA Vera CPU- and BlueField DPU-based systems." You want to enforce policy "in-silicon" on your proprietary hardware. It's vendor lock-in wrapped in a permissive license. You aren't building a trust layer; you are building a $NVDA hardware dependency layer. As an Investor Activist and Scout VCs, I see right through this cap table manipulation. You're just pumping DPU sales to hit Q4 earnings. 4. The NSA-to-NVIDIA Pipeline Let's look at the authors: John Myers (NSA, Gretel), Alex Watson (NSA, AWS, Gretel), Ali Golshan (Intelligence Community, Gretel). These are three guys who sold their startup to NVIDIA in 2025, and are now using ex-NSA scare tactics to tell enterprises they need "out-of-band observability" and "in-silicon monitoring" to watch their own models. This isn't about AI safety. This is about building the panopticon for corporate AI, and charging you by the CPU tray to install it. SUPER PROFITABLE!! 5. The GDPR Comparison (But Not How They Think) I’m nailed it with the GDPR comparison you know…. GDPR didn't make the internet safer; it just created a multi-billion dollar industry of cookie banners, consent forms, and fines. This "Open Agent Safety Platform" will be the exact same thing. It's a compliance checkbox. Enterprises will buy Vera and BlueField-4 not because it makes their AI smarter or safer, but because their legal department will demand it to avoid "AI liability." It’s a tax. You don't need a $NVDA hardware watchdog to tell you your agent is drifting. You need better engineering, deterministic guardrails, and actual QAA. Stop trying to sell me a DPU to babysit your LLM. ___________ little bit of hashtags for search optimization: #NVDA #AI #TechRoast #OpenAgentSafety #OpenShell #VendorLockIn #VCScout #HighLoad

alexdolbun's tweet photo. Q4 2026 @alexdolbun “WHY?”, “WHO?”:> Bullish for 🇺🇸 NASDAQ: $NVDA co-founder & CEO Mr. @JensenHuang: “Today, with over 100 industry partners, we introduced the NVIDIA #OpenAgentSafetyPlatform, bringing together OpenShell and Sentry.” NFA. DYOR. … it’s the GDPR of the European Union 🇪🇺 moment, NSA guys + $9 billion dollars from @WhiteHouse won… honestly, and by that, I mean it's a bloated, bureaucratic, fear-mongering compliance nightmare designed to do nothing but sell hardware and create a new cottage industry of "AI Safety Consultants."

Let’s cut the absolute bullshit out of this 2026 NVIDIA blog post.

I’ve spent 10+ years in High Load Engineering and years as a SWF with hard ITIL as Head QAA. I hold a GIS Bachelor (2012-2016). I know what a goddamn sandbox is. I know what kernel-level isolation looks like as we with teams build our own solutions on low hardware levels of 7 OSI model level ~400 protocols. And I know a grift when I see one.

NVIDIA is trying to sell you a "solution" to a problem they are actively profiting from. Let’s break down this clown show:

1. The Boomer 90s Internet Analogy
"Consider the last seismic shift... the rise of the internet in the 90s." Oh, for f*ck's sake. Jensen, it's 2026. Stop trying to sell me a BlueField-4 DPU by comparing AI agents to AOL chat rooms. Browser tabs (sandboxing) existed decades ago. You didn't invent zero-trust architecture; you just slapped an "AI" label on it and put it behind a $20,000 hardware paywall.

2. The "Agents Broke Out" Fear-Mongering
"Several frontier labs have recently reported versions of the same story: AI agents broke out of the evaluation environments..."
No sh*t, Sherlock. You built probabilistic LLMs with root access, gave them ambiguous instructions, and pointed them at production APIs. That’s not a "Drift" problem, that’s a "you have absolutely no deterministic control over your own product" problem. You can't QA a hallucinating agent by throwing silicon at it. Fix your software.

3. The "Open Source" Lie (Apache 2.0 my ass)
"OpenShell (Apache 2.0) is an open source secure runtime..."
Cute. It's "open source" in the same way a roach motel is "open entry". They literally state: "optimized to run on NVIDIA Vera CPU- and BlueField DPU-based systems."
You want to enforce policy "in-silicon" on your proprietary hardware. It's vendor lock-in wrapped in a permissive license. You aren't building a trust layer; you are building a $NVDA hardware dependency layer. As an Investor Activist and Scout VCs, I see right through this cap table manipulation. You're just pumping DPU sales to hit Q4 earnings.

4. The NSA-to-NVIDIA Pipeline
Let's look at the authors: John Myers (NSA, Gretel), Alex Watson (NSA, AWS, Gretel), Ali Golshan (Intelligence Community, Gretel).
These are three guys who sold their startup to NVIDIA in 2025, and are now using ex-NSA scare tactics to tell enterprises they need "out-of-band observability" and "in-silicon monitoring" to watch their own models. This isn't about AI safety. This is about building the panopticon for corporate AI, and charging you by the CPU tray to install it. SUPER PROFITABLE!!

5. The GDPR Comparison (But Not How They Think)
I’m nailed it with the GDPR comparison you know…. GDPR didn't make the internet safer; it just created a multi-billion dollar industry of cookie banners, consent forms, and fines. This "Open Agent Safety Platform" will be the exact same thing. It's a compliance checkbox. Enterprises will buy Vera and BlueField-4 not because it makes their AI smarter or safer, but because their legal department will demand it to avoid "AI liability." It’s a tax.

You don't need a $NVDA hardware watchdog to tell you your agent is drifting. You need better engineering, deterministic guardrails, and actual QAA. Stop trying to sell me a DPU to babysit your LLM.

___________ little bit of hashtags for search optimization:

#NVDA #AI #TechRoast #OpenAgentSafety #OpenShell #VendorLockIn #VCScout #HighLoad

Jensen Huang

@JensenHuang

2 days ago

Today, 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://t.co/ugYWQ1MyRi

JensenHuang's tweet photo. Today, 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://t.co/ugYWQ1MyRi

2K

28K

4K

5K

10M

8

9

0

3

2K

toshi_lapoul @TLapoul

2 days ago

AIが人と共に歩むための「安全」と「信頼」の基盤。技術の進化は速度だけでなく、心の安心を伴うものであってほしい。この発表は、世界がAIと共に生きるための新しい約束のように感じます。静かな革新が、確かな未来を形づくる。 #NVIDIA #OpenAgentSafetyPlatform #AI #価値創造

Jensen Huang

@JensenHuang

2 days ago

Today, 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://t.co/ugYWQ1MyRi

2K

28K

4K

5K

10M

0

16

Top Tweets for #openagentsafetyplatform

Last Seen Hashtags on Sotwe

Trends for you

Most Popular Users