Published on November 8, 2023, 4:16 pm
Since its launch, Microsoft’s Bing Chat has been making waves in the news, though not all of the attention has been positive. Recently, a new issue has come to light regarding the chatbot’s tendency to direct users towards malware-infected websites.
Antivirus firm Malwarebytes discovered this flaw and shared their findings in a blog post. Apparently, Bing Chat is displaying malware advertisements that redirect users to malicious websites instead of filtering them out. When using the chatbot, users can ask for information or links related to various topics. However, since its initial release, Microsoft has been inserting ads into these links similar to how Google places ads above search results.
The problem arises when bad actors purchase ads that lead unsuspecting users to deceptive websites posing as legitimate destinations. If users are not careful, they can easily fall victim to this bait-and-switch tactic. Malwarebytes highlights how this works by providing an example of asking Bing Chat to download a popular IP scanning app used by system admins. While Bing Chat displays a link supposedly directing users to the app’s official website, hovering over the link reveals both the real website and a malicious advertisement placed directly above it.
Without close scrutiny or familiarity with the official website address, users may unknowingly click on the deceptive link rather than accessing the genuine site. Further investigation by Malwarebytes revealed that this fake website redirects visitors to another site with a very similar web address to the real app’s official URL. Once there, users are prompted to download malware that could harm their computers.
This incident sheds light on the need for Microsoft to enhance its protection measures against malicious ads served through Bing Chat. For now, it is advisable for users to exercise caution when clicking on links provided by Bing Chat. It might be safer to stick with traditional search engines and consider installing ad blockers as an added defense against encountering such malicious adverts.
In other AI-related news, Zoom video-calling app recently introduced its own “AI Companion” assistant. This tool, developed in collaboration with OpenAI and Meta, incorporates artificial intelligence (AI) and large language models (LLMs). The AI Companion aims to help users catch up on missed meetings and respond quickly to chat messages while enhancing productivity, connectivity, and collaboration.
Furthermore, OpenAI’s ChatGPT developer discusses the potential roles its large language model (LLM), GPT-4, may assume in the future. In a recent blog post from OpenAI, it is suggested that GPT-4 could function as an online moderator capable of monitoring forums and social networks for objectionable content. By leveraging AI instead of human moderators, OpenAI claims that policy changes can be implemented rapidly, reducing the time taken from months to hours. Additionally, GPT-4’s ability to interpret complex content policy documentation swiftly enables consistent labeling updates in response to policy changes.
While Macs have long held a reputation for being less susceptible to malware than Windows devices, advancements in AI tools like ChatGPT are potentially challenging this notion. In fact, even the FBI has raised concerns about the implications of AI-driven cybersecurity breaches. As a response to this evolving landscape, software developer Macpaw created Moonlock, its dedicated cybersecurity division focused on combating Mac malware. We spoke with Oleg Stukalenko, Lead Product Manager at Moonlock, seeking insights into whether there is indeed a rise in Mac malware threats and how AI tools like ChatGPT may offer hackers an advantage over everyday users.
For more news on tech advancements and product reviews across various fields, Digital Trends is a reliable source providing readers with current updates and unique glimpses into the world of technology.