Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Nvidia NemoClaw

github.com

Post reply on HN
Print view View original

211–220 of 291 posts

Re: Nvidia NemoClaw

#211

Post by heavyset_go »

observationist wrote: post #133
Netcob wrote: post #65

Am I missing something? Why is everyone talking about sandboxes when it comes to OpenClaw? To me it's like giving your dog a stack of important documents, then being worried he might eat them, so you put the dog in a crate, together with the documents. I thought the whole problem with that idea was that in order for the agent to be useful, you have to connect it to your calendar, your e-mail provider and other servic…

What makes it even better is that these dogs are like Malinois. If they want to get into something, they will; people have had their entire network compromised by bots they left running overnight, and any important information like account logins and so on runs the risk of being misused. It's one thing to sandbox, maybe give the bot a temporary, limited $100 card or account to go perform a specific task, but there's…

Agent psychosis is just as prevalent as AI psychosis

Re: Nvidia NemoClaw

#212

Post by sfdlkj3jk342a »

Someone1234 wrote: post #76
Netcob wrote: post #65

Am I missing something? Why is everyone talking about sandboxes when it comes to OpenClaw? To me it's like giving your dog a stack of important documents, then being worried he might eat them, so you put the dog in a crate, together with the documents. I thought the whole problem with that idea was that in order for the agent to be useful, you have to connect it to your calendar, your e-mail provider and other servic…

I think the point you're making is fully correct, so consider this a devil's advocate argument... People claim, you can use Claw-agents more safely while getting some of the benefits, by essentially proxying your services. For example on Gmail people are creating a new Google accounts, forwarding email via rule, and adding access to their calendar via Google's Family Sharing. This allows the Claw agent to read email,…

> I think insecure/dangerous Claw-agents could be useful but cannot be made safe

Isn't it a question of when they will be "safe enough"? Many people already have human personal assistants, who have access to many sensitive details of their personal lives. The risk-reward is deemed worth it for some, despite the non-zero chance that a person with that access will make mistakes or become malicious.

It seems very similar to the point when automated driving becomes safe enough to replace most human drivers. The risks of AI taking over are different than the risks of humans remaining in control, but at some point I think most will judge the AI risks to have a better tradeoff.

Re: Nvidia NemoClaw

#213

Post by heavyset_go »

ttsalami wrote: post #39
bigfishrunning wrote: post #22

Earlier quoted context omitted.

> OpenClaw lets people live a bit dangerously, but fundamentally gives them something that they actually wanted. They wanted it so badly that they're willing to take what seem like insane risks to get it. For the first time in my career I feel so incredibly behind on this: What is open claw giving people that they want so badly? It just seems like Russian Roulette, I honestly don't see the upside

I can give you, as an example, what is driving me towards trying it. I work as a contractor for 2 companies, not out of necessity, but greed. I also have a personal project with a friend that is dangerously close to becoming a business that needs attention. I also have other responsibilities and believe it or not - friends. Also the ADHD on top of that. I yearn for a personal assistant. Something or somebody that wil…

You will learn to ignore the bot like everything else.

You're looking for a technical solution to a problem that is not technical. Saying this as someone who is similar to you.

Re: Nvidia NemoClaw

#214

Post by mark_l_watson »

Netcob wrote: post #65

Am I missing something? Why is everyone talking about sandboxes when it comes to OpenClaw? To me it's like giving your dog a stack of important documents, then being worried he might eat them, so you put the dog in a crate, together with the documents. I thought the whole problem with that idea was that in order for the agent to be useful, you have to connect it to your calendar, your e-mail provider and other servic…

I agree, but would like to go further: I won’t run OpenClaw type systems because of security and privacy reasons. Although I dislike making tech giants even more powerful, it seems safer to choose your primary productivity platform (Google Workplace, Apple ecosystem, or Microsoft) and wait for them to implement hopefully safer OpenClaw type systems just for their ecosystems and take advantage of centralized security, payment systems, access to platform cloud files, etc. Note: I use ProtonMail, prefer using local models, etc. so when I talk about going all-in on one huge platform I am not talking about anything I want to do in the foreseeable future.

Re: Nvidia NemoClaw

#215

Post by Yokohiii »

Netcob wrote: post #65

Am I missing something? Why is everyone talking about sandboxes when it comes to OpenClaw? To me it's like giving your dog a stack of important documents, then being worried he might eat them, so you put the dog in a crate, together with the documents. I thought the whole problem with that idea was that in order for the agent to be useful, you have to connect it to your calendar, your e-mail provider and other servic…

Neither NVIDIA or OpenClaw bros care about security at this point. NVIDIA of course wants to fuel the hype train and will proudly point to this, adding 0.1% security to an 2000% insecurity. Most bros wont even mind, produce insecure crap at light speed and never look back. It's probably just there to trick silly non tech corps into this junk.

Re: Nvidia NemoClaw

#216

Post by ori_b »

sfdlkj3jk342a wrote: post #212
Someone1234 wrote: post #76

Earlier quoted context omitted.

I think the point you're making is fully correct, so consider this a devil's advocate argument... People claim, you can use Claw-agents more safely while getting some of the benefits, by essentially proxying your services. For example on Gmail people are creating a new Google accounts, forwarding email via rule, and adding access to their calendar via Google's Family Sharing. This allows the Claw agent to read email,…

> I think insecure/dangerous Claw-agents could be useful but cannot be made safe Isn't it a question of when they will be "safe enough"? Many people already have human personal assistants, who have access to many sensitive details of their personal lives. The risk-reward is deemed worth it for some, despite the non-zero chance that a person with that access will make mistakes or become malicious. It seems very simila…

A personal assistant is responsible for their own gross negligence and malicious actions. I can take them to court to attempt to recover damages.

When Anthropic is willing enough to stand behind their agents strongly enough to accept liability for their actions, we can talk.

Re: Nvidia NemoClaw

#217

Post by clawbridge »

the answer isn't sandbox everything, it's knowing which steps need AI judgment and which should be deterministic code. I lean towards the latter as much as possible

Re: Nvidia NemoClaw

#218

Post by simple10 »

Netcob wrote: post #65

Am I missing something? Why is everyone talking about sandboxes when it comes to OpenClaw? To me it's like giving your dog a stack of important documents, then being worried he might eat them, so you put the dog in a crate, together with the documents. I thought the whole problem with that idea was that in order for the agent to be useful, you have to connect it to your calendar, your e-mail provider and other servic…

Yeah, it's wild. I spent several weeks nearly full time on a deep dive of claw architecture & security.

The short of it - OpenClaw sandboxes are useful for controlling what sub-agents can do, and what they have access to. But it's a security nightmare.

During config experiments, I got hit with a $20 Anthropic API charge from one request that ran amuck. Misconfigured security sandbox issue resulted in Opus getting crazy creative to find workarounds. 130 tool calls and several million tokens later... it was able to escape the sandbox. It used a mix of dom-to-image sending pixels through the context window, then writing scripts in various sandboxes to piece together a full jailbreak. And I wasn't even running a security test - it was just a simple chat request that ran into sandbox firewall issues.

Currently, I use sandboxes to control which agents (i.e. which system prompts) have access to different tools and data. It's useful, but tricky.

Re: Nvidia NemoClaw

#219

Post by simple10 »

NemoClaw is mostly a trojan horse of sorts to get corporate OpenClaw users quickly ported over to Nvidia's inference cloud.

It's a neat piece of architecture - the OpenShell piece that does the security sandboxing. Gives a lot more granular control over exec and network egress calls. Docker doesn't provide this out of the box.

But NemoClaw is pre-configured to intercept all OpenClaw LLM requests and proxy them to Nvidia's inference cloud. That's kinda the whole point of them releasing it.

I can be modified to allow for other providers, but at the time of launch, there was no mention of how to do this in their docs. Kinda a brilliant marketing move on their part.

Post reply on HN