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

221–230 of 433 posts

Re: Claude for Chrome

#221

Post by zaphirplane »

mark242 wrote: post #209
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.

What ! 1 in 10 successfully phished is ok ? 1 in 10 page views. That has to approach 100% success rate over a week say month of browsing the web with targeted ads and/or link farms to get the page click

Re: Claude for Chrome

#222

Post by whatevertrevor »

asdff wrote: post #200
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.

> "We conducted extensive adversarial prompt injection testing, evaluating 123 test cases representing 29 different attack scenarios. " Doesn't this seem like a remarkably small set of tests? And the fact that it took this testing to realize that prompt injection and giving the reigns to the AI agent is dangerous strikes me as strange that this wasn't anticipated while building the tool in the first place, before it…

I wonder how this will even fare in the review process, or if the big AI players will get a free pass here. My intuition says that it's a risk that Google/Chrome absolutely don't want to own, it will be curious to see how "Agentic" AI gets deployed in browsers from a liability fallout perspective.

Re: Claude for Chrome

#223

Post by sega_sai »

It is clear that a lot of things: programming languages, websites and others will have to be adapted to be easier to use for LLMs. Now they are optimized for humans, but I think very soon they will be optimized for LLMs instead.

Re: Claude for Chrome

#224

Post by onesociety2022 »

lemonberry wrote: post #93

I love Claude via the website interface. I can't wait to try Claude Code. Once I have a separate computer with none of my personal information or files on it I'm going to use the heck out of it. I'd probably even install Claude for Chrome on it.

If you don't give the agent access to any of your personal information, how useful is it really going to be? The agent can only help you with tasks that can be accomplished by browsing the web anonymously.

Re: Claude for Chrome

#225

Post by asdff »

whatevertrevor wrote: post #222
asdff wrote: post #200

Earlier quoted context omitted.

> "We conducted extensive adversarial prompt injection testing, evaluating 123 test cases representing 29 different attack scenarios. " Doesn't this seem like a remarkably small set of tests? And the fact that it took this testing to realize that prompt injection and giving the reigns to the AI agent is dangerous strikes me as strange that this wasn't anticipated while building the tool in the first place, before it…

I wonder how this will even fare in the review process, or if the big AI players will get a free pass here. My intuition says that it's a risk that Google/Chrome absolutely don't want to own, it will be curious to see how "Agentic" AI gets deployed in browsers from a liability fallout perspective.

Probably no liability considering that is how other phishing attempts are viewed.

Re: Claude for Chrome

#226

Post by miguelspizza »

onesociety2022 wrote: post #212
chatmasta wrote: post #195

Earlier quoted context omitted.

I suspect this kind of framework will be adopted by websites with income streams that are not dependent on human attention (i.e. advertising revenue, mostly). They have no reason to resist LLM browser agents. But if they’re in the business of selling ads to human eyeballs, expect resistance. Maybe the AI companies will find a way to resell the user’s attention to the website, e.g. “you let us browse your site with an…

Even the websites whose primary source of revenue is not ad impressions might be resistant to let the agents be the primary interface through which users interact with their service. Instacart currently seems to be very happy to let ChatGPT Operator use its website to place an order ( https://www.instacart.com/company/updates/ordering-groceries... ) [1]. But what happens when the primary interface for shopping with I…

The side-panel browser agent is a good middle ground to this issue. The user is still there looking at the website via their own browser session, the AI just has access to the specific functionality which the website wants to expose to it. The human can take over or stop the AI if things are going south.

Re: Claude for Chrome

#227

Post by padolsey »

This is a huge shame. Browsers are one of the more ideal sandboxing barriers. The likes of Chrome and Firefox could have worked with OSs and AI labs to ensure a more robust system of mitigations were in place. Setting legitimizing precedent and making such things official will not end well.

Re: Claude for Chrome

#228

Post by miguelspizza »

chatmasta wrote: post #195
miguelspizza wrote: post #176

Earlier quoted context omitted.

> It's clear to me that the tech just isn't there yet. Totally agree. This was the thesis behind MCP-B (now WebMCP https://github.com/MiguelsPizza/WebMCP ) HN Post: https://news.ycombinator.com/item?id=44515403 DOM and visual parsing are dead ends for browser automation. Not saying models are bad; they are great. The web is just not designed for them at all. It's designed for humans, and humans, dare I say, are prett…

I suspect this kind of framework will be adopted by websites with income streams that are not dependent on human attention (i.e. advertising revenue, mostly). They have no reason to resist LLM browser agents. But if they’re in the business of selling ads to human eyeballs, expect resistance. Maybe the AI companies will find a way to resell the user’s attention to the website, e.g. “you let us browse your site with an…

The Primary client for WebMCP enabled websites is a chrome extension like Claude Chrome. So the human is still there in the loop looking at the screen. MCP also supports things like elicitation so the website could stop the model and request human input/attention

Re: Claude for Chrome

#229

Post by asdff »

jascha_eng wrote: post #220
asdff wrote: post #210

Earlier quoted context omitted.

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?

Oh the tools are hand coded (or rather built with Claude Code) but the agent can call them to control the browser. Imagine a prompt like this: You are a research agent your goal is to figure out this companies tech stack: - Company Name Your available tools are: - navigate_to_url: use this to load a page e.g. use google or bing to search for the company site It will return the page content as well as a list of availa…

I still don't understand what the llm does. One could do this with a few lines of curl and a list of tools to query against.

Re: Claude for Chrome

#230

Post by sarreph »

Funny timing. On the weekend I built something that provides screenshots only (no KBAM controls) from Chrome to Claude Code, so that it has visual context on what you're building. Keen to see how the two way control works out with Claude-in-Chrome though!

For anyone interested it's called MagicEyes (https://github.com/rorz/MagicEyes) and it's in alpha!

Post reply on HN