Copyright news18
AI web browsers like ChatGPT Atlas, Comet and even Google’s Chrome are exciting and many people are eager to switch to the new era of the internet. However, there is a strong reason for users to reconsider their choice or even be careful about using these AI browsers. Security experts are worried that any level of demand for these browsers will push hackers to bypass its issues that can make it easy to steal data and valuable information. The amount of features and ease of using an AI browser makes it a compelling option for millions, and as many say, enables multi-tasking at a whole new level. But that convenience could become costly and experts have raised the red flag over this situation. AI Browsers Are Cool But Not Safe The main danger facing AI browsers like ChatGPT Atlas and Comet is called prompt injections. We all know that AI chatbots respond to prompts, be it for creating an image, ordering groceries or even asking the AI to scan through a document. For instance, when you make an AI shop for your grocery, it is likely to include the payment details as well, and these bad actors could use prompt injections to make the AI model reveal those numbers. These injections can also attack users who simply get the AI to summarise a malicious web page, it is that simple. These attacks seem inevitable with the growth of AI browsers and hackers will be licking their lips knowing the power of AI and how the technology can be manipulated with prompts that are injected into web pages to deceive the AI model and make it do things the wrong way. Perplexity’s Comet has already been scrutinised for its security issues which leaves the threat of prompt injections wide open. We have seen AI hallucinate and dish out information that does not even exist on the internet. The chances of AI browsers going rogue are very possible because of these instances and platforms like Brave, investigating these browsers have warned users about the realities of these apps. “As users grow comfortable with AI browsers and begin trusting them with sensitive data in logged in sessions—such as banking, healthcare, and other critical websites—the risks multiply,” the company mentioned in a post. They Need Your Data Have you ever thought how these AI browsers can handle your tasks and automate certain activities? Well, they need a lot of access and your data to make it work. Not just the regular email ID, phone number and contact list but a lot more elaborate. This includes reading through your emails, enabling plugin access for third-party tasks, your browsing history, your online activities and more. Be Smart And Cautious The dangers of AI browsers are evident and some proof-of-concept demos show the severity of its impact. But that does not mean people should be completely against using new technology. We’ve seen the leap happen in every generation and the risks exist in every aspect. People just need to be careful how they use these AI browsers, how much data they are willing to share with the AI model and what all tasks they really want automated.
