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

Claude for Chrome

anthropic.com

Post reply on HN
Print view View original

201–210 of 433 posts

Re: Claude for Chrome

#201

Post by asdff »

erickhill wrote: post #193
stusmall wrote: post #181

It's wild to see an AI company put out a press release that is basically "hey, you kids wanna see a loaded gun?" Normally all their public coms are so full of optimism and salesmanship around the potential. They are fully aware of how dangerous this is.

Seems to be trying to explain why the rollout is going to be very focused and rather small at first so they can build the proper safeguards. But it is a surprising read, you're absolutely right.

Letting their beta testers get pwned is an interesting opsec strategy indeed.

Re: Claude for Chrome

#203

Post by ankit219 »

stusmall wrote: post #181

It's wild to see an AI company put out a press release that is basically "hey, you kids wanna see a loaded gun?" Normally all their public coms are so full of optimism and salesmanship around the potential. They are fully aware of how dangerous this is.

This is what they need for the next generation of models. The key line is:

> We view browser-using AI as inevitable: so much work happens in browsers that giving Claude the ability to see what you're looking at, click buttons, and fill forms will make it substantially more useful.

A lot of this can be done by building a bunch of custom environments at training time, but only a limited number of usecases can be handled that way. They don't need the entire data, they still need the kind of tasks real world users would ask them to do.

Hence, the press release pretty much saying that they think it's unsafe, they don't have any clue how to make it safe without trying it out, and they would only want a limited number of people to try it out. Give their stature, it's good to do it publicly instead of how Google does it with trusted testers or Openai does it with select customers.

Re: Claude for Chrome

#204

Post by asdff »

parsabg wrote: post #126

I built a very similar extension [1] a couple of months ago that supports a wide range of models, including Claude, and enables them to take control of a user's browser using tools for mouse and keyboard actions, observation, etc. It's a fun little project to look at to understand how this type of thing works. It's clear to me that the tech just isn't there yet. The information density of a web page with standard rep…

It is kind of funny how the systems are set up where there often is dense and queryable information out there already for a lot of these tasks, but these are ignored in favor of the difficult challenge of brute forcing the human consumer facing ui instead of some existing api that is designed to be machine readable already. E.g. booking flights. Travel agents use software that queries all the airlines ticket inventory to return flight information to you the consumer. The issue of booking a flight is theoretically solved already by virtue of these APIs that already exist to do just that. But for AI agents this is now a stumbling block because it would presumably take a little bit of time to craft out a rule to cover this edge case and return far more accurate information and results. Consumers with no alternative don't know what they are missing so there is no incentive to improve this.

Re: Claude for Chrome

#205

Post by clutchdude »

throwaway13337 wrote: post #191
clutchdude wrote: post #175

Earlier quoted context omitted.

> If you could accomplish your task without the busywork, why wouldn’t you? There's taking away the busywork such as hand washing every dish and instead using a dishwasher. Then there is this where, rather than have any dishes, a cadre of robots comes by and drops a morsel of food in your mouth for every bite you take.

Does your analogy mean that you'd like to stop someone from owning that cadre of robots? Or is this just a personal preference? You can have your dishwasher and I'll take the robots. And we can both be happy.

And therein is the problem - if your robots take up so many resources I can't have my dishwasher, is that your right? Is your right to being happy more important than others?

Re: Claude for Chrome

#206

Post by devmor »

throwaway13337 wrote: post #191
clutchdude wrote: post #175

Earlier quoted context omitted.

> If you could accomplish your task without the busywork, why wouldn’t you? There's taking away the busywork such as hand washing every dish and instead using a dishwasher. Then there is this where, rather than have any dishes, a cadre of robots comes by and drops a morsel of food in your mouth for every bite you take.

Does your analogy mean that you'd like to stop someone from owning that cadre of robots? Or is this just a personal preference? You can have your dishwasher and I'll take the robots. And we can both be happy.

A more detailed analogy would be if you owning the robots meant that all food is now packaged for robots instead of humans, increasing the personal labor cost of obtaining and preparing food as well as inflating the cost of dinnerware exponentially, while driving up my power bill to cover the cost of expanding infrastructure to power your robots.

In that case, I certainly am against you owning the robots and view your desire for them as a direct and immediate threat against my well being.

Re: Claude for Chrome

#207

Post by adam_arthur »

parsabg wrote: post #126

I built a very similar extension [1] a couple of months ago that supports a wide range of models, including Claude, and enables them to take control of a user's browser using tools for mouse and keyboard actions, observation, etc. It's a fun little project to look at to understand how this type of thing works. It's clear to me that the tech just isn't there yet. The information density of a web page with standard rep…

The LLM should not be seeing the raw DOM in its context window, but a highly simplified and compact version of it.

In general LLMs perform worse both when the context is larger and also when the context is less information dense.

To achieve good performance, all input to the prompt must be made as compact and information dense as possible.

I built a similar tool as well, but for automating generation of E2E browser tests.

Further, you can have sub-LLMs help with compacting aspects of the context prior to handing it off to the main LLM. (Note: it's important that, by design, HTML selectors cannot be hallucinated)

Modern LLMs are absolutely capable of interpreting web pages proficiently if implemented well.

That being said, things like this Claude product seem to be fundamentally poorly designed from both a security and general approach perspective and I don't agree at all that prompt engineering is remotely the right way to remediate this.

There are so many companies pushing out junk products where the AI is just handling the wrong part of the loop and pulls in far too much context to perform well.

Re: Claude for Chrome

#208

Post by hodgehog11 »

stusmall wrote: post #181

It's wild to see an AI company put out a press release that is basically "hey, you kids wanna see a loaded gun?" Normally all their public coms are so full of optimism and salesmanship around the potential. They are fully aware of how dangerous this is.

I noticed this with the OpenAI presentation for GPT-5 too; they just dove straight in to some of the less ethical applications (writing a eulogy, medical advice, etc.). But while the OpenAI presentation felt more like kids playing with a loaded gun, this feels more like inevitability: "we're already heading down this path anyway, so it may as well be us that does it right".

Re: Claude for Chrome

#209

Post by mark242 »

biggestfan wrote: post #8

According to their own blog post, even after mitigations, the model still has an 11% attack success rate. There's still no way I would feel comfortable giving this access to my main browser. I'm glad they're sticking to a very limited rollout for now. (Sidenote, why is this page so broken? Almost everything is hidden.)

11% success rate for what is effectively a spear-phishing attempt isn't that terrible and tbh it'll be easier to train Claude not to get tricked than it is to train eg my parents.

Re: Claude for Chrome

#210

Post by asdff »

jascha_eng wrote: post #146
aliljet wrote: post #22

Having played a LOT with browser use, playwright, and puppeteer (all via MCP integrations and pythonic test cases), it's incredibly clear how quickly Claude (in particular) loses the thread as it starts to interact with the browser. There's a TON of visual and contextual information that just vanishes as you begin to do anything particularly complex. In my experience, repeatedly forcing new context windows between sc…

I have built a custom "deep research" internally that uses puppeteer to find business information, tech stack and other information about a company for our sales team. My experience was that giving the LLM a very limited set of tools and no screenshots worked pretty damn well. Tbf for my use case I don't need more interactivity than navigate_to_url and click_link. Each tool returning a text version of the page and th…

Seems navigate_to_url and click_link would be solved with just a script running puppeteer vs having an llm craft a puppeteer script to hopefully do this simple action reliably? What is the great advantage with the llm tooling in this case?
Post reply on HN