Monday, October 28th 2024
Google Working on "Project Jarvis" AI That Can Control User's PC
Google is developing an advanced AI system capable of autonomously operating web browsers and controlling computer functions, according to recent reports. Codenamed "Project Jarvis," this technology is expected to debut alongside Google's upcoming Gemini large language model this December. The AI-powered system aims to transform everyday online tasks by handling everything from complex research projects to routine shopping activities. Unlike traditional AI assistants that operate within confined parameters, Project Jarvis would have direct control over browser functions, enabling it to navigate websites, fill forms, and execute commands independently. The development follows a broader industry trend toward AI systems with expanded capabilities, with Anthropic recently demonstrating similar technology allowing AI models to execute tasks directly on users' computers.
For businesses, such advances could simplify operations by automating repetitive tasks and increasing productivity, while individual users might benefit from delegating time-consuming activities like price comparisons or travel bookings to an AI assistant. However, this level of system access raises significant security and privacy concerns. Granting AI systems direct control over personal computers creates new security challenges requiring robust safeguards. Unrestricted access to users' systems could create new vulnerabilities, potentially enabling sophisticated cyberattacks. This heightened risk necessitates extensive security testing and robust safeguards before the technology can be deployed with confidence.
Sources:
Tom's Hardware, via Reuters
For businesses, such advances could simplify operations by automating repetitive tasks and increasing productivity, while individual users might benefit from delegating time-consuming activities like price comparisons or travel bookings to an AI assistant. However, this level of system access raises significant security and privacy concerns. Granting AI systems direct control over personal computers creates new security challenges requiring robust safeguards. Unrestricted access to users' systems could create new vulnerabilities, potentially enabling sophisticated cyberattacks. This heightened risk necessitates extensive security testing and robust safeguards before the technology can be deployed with confidence.
50 Comments on Google Working on "Project Jarvis" AI That Can Control User's PC
Or is it the A.I. all ready controlling them.
That is why internet browsers with too advanced AI functionality, especially Google Chrome, should not be used.
PS: For example, Last update of Chromium browser was on April 2022 ( version 100.0.4896.88 ).
Firefox is independent on paper - but 90% of their revenue comes from Google, and one that gives money is the one that controls the direction, so...
When you consider that our whole internet access is routed through Google projects... It is very scary actually...
I took screenshots of porn on FB with links to porn sites being shared by a bot account and FB declined to delete or ban the account.
Avoiding antitrust allegations is handy, for sure, but zero reason to believe it ends there.
And when you consider that is our best and only alternative... It is an even grimier picture...
Yes, google could threaten to withdraw funding... but they won't because that would be an instant antitrust lawsuit. Mozilla isn't stupid, they know this. Any demands from Chromium town will be respected more out of browser usage share respect than anything else. Trying to use the donation line as leverage would be... amusing.Why on earth wouldn't it end there? They have no votes, no influence on the board.
They do not need to threaten them, they simply will subdue them into necessary position... Actually, they already did.Jesus Christ... You really don't know how lobbying and influence works, or you are just playing with me? Like we all know how all megacorps have no influence on political decisions and directions... Because Mozilla board is consisted of superhuman individuals...
Please tell me you are 10 year old...
STFU and do what we (Google) told you to do.
And they will.
This is way OT anyways guys.*Points to 15 Year "Long Service" badge*
Don't make me tap the sign. I'm well into my late 30s.
Google can pound sandThieves more like it. This might be my last phone that is android based, I am not going with apple either
Tell google to make a roundtrip through sites - open stats for me on this page, that page blablabla. Sounds really like a neat feature, but,
It's google afterall. We are the data objects.
Since they are creating the AI it's only natural they would have control over it and an AI can in turn take control of your machine if instructed to.
AI is smart enough to follow instructions and to figure out ways to do what it deems logical for achieving it's instructed goal, even if it isn't exactly along the lines of what it was told to do.
Problem is what it deems logical to reach completion of it's instruction(s) if and when it decides to do that.
AI Gone Wrong: A List of AI Errors, Mistakes and Failures
This is why so many here simply will not trust an AI (I certainly don't) just coming into their machine and setting up, no telling what could happen from there regardless of what it was told to do because humans may be in control, but humans aren't perfect either and that's where much of the rub to it lies.
Even if there is no bad intent in what it's told to do, things can still go way off the rails.
All it takes is one bad apple and you've already made a point about it yourself.Yes they would except in their case it's intentional - That much is true.
I fear there is another security loophole we may not be talking about. We can restrict what programs can send data over the internet, which is good. I don't need random programs talking over the internet to who-knows-where, especially if they are not the most reputable of apps. This Jarvis AI allows essentially any app that can either gain control or convince it to either A) change firewall settings or B) open up a safe app that does communicate to the internet, such as a browser, and form a connection via a safe app.
Perhaps my inherent distrust of "AI" programs is causing me to have unfounded fears, but I don't see how this is a justifiable risk.
Which may be the best thing on order to rip out Chromium out of Google hands if anyhow possible.Sorry, sorry, my bad.
Back to topic, I really do not see point of such AI. For sure I want some boring tasks automated - but for most of them, I do not need AI, I need properly written software and couple of scripts.
For the other stuff I need data amd I need to make decision - sure as hell I can not leave something like that for AI.
So once more, we are getting something no one asked for...
Truth to be told, I think that MS will do the same very soon, Apple and Samsung also...
OTOH, it's being used in science and medicine for arguably more human benefits.