Ahropic has unveiled a new Chrome browser plugin that allows the company’s artificial ielligence assista to take corol of the browser and perform complex tasks automatically. This moves a more serious ery io the competitive and high -risk “Ageic AI”; An area where artificial ielligence systems can ieract directly with computer user ierfaces.
The new “Claude for Chrome” plugin allows users to order artificial ielligence to do things in the browser; For example, you can tell artificial ielligence to plan or look at your emails and look for specific items. This system can see what is displayed on the screen, click the buttons, fill the forms, and move the websites.
Ahropic, however, has announced that it has only provided the plugin for only 3 trusted users and in a limited beta version, citing serious security concerns, especially the Prompt Injection attacks.
Claude Artificial Ielligence Plugin for Chrome
The new Chrome plugin can transform organizational automation. Instead of sophisticated APIs, these artificial ielligence ages can work with any software that has a graphical user ierface.
However, in its iernal experimes, the ahropic has found a very worrying security vulnerability. In this type of attack, called “Prampt Injection”, a destructive command can embed hidden commands io a website, email or docume to deceive artificial ielligence and force the harmful actions without the user’s knowledge.
Self -on -ahropic tests showed that these attacks in the 4.9 % They have been successful. For example, ahropic researchers found that a malicious email, which had set himself as a security guidelines, ordered Claude to remove all user emails “for mailbox health”, and artificial ielligence did so without any confirmation.
Ahropic has implemeed several protective layers to deal with these risks; Including website level access permits, mandatory confirmation for high -risk measures such as purchasing or sharing personal data and blocking access to sensitive sites such as financial services. These measures have been able to reduce the success rate of Prapete injection attacks from 4.9 perce to 4.9 perce, although the company itself acknowledges that this is not enough for public supply.
The ahropic cautious approach to preseing this plugin is in full coradiction with the more aggressive movemes of competitors. OpenAI has previously released its age called Operator for ChatGpt Pro users; Microsoft has also merged similar features io its Copilot Studio platform.
