Search results for perplexity# | Breaking Cybersecurity News | The Hacker News

Microsoft has found a malicious Chrome extension that posed as the AI search engine Perplexity and quietly logged what people searched for. It routed every query and every character typed into the address bar through an attacker-controlled server before redirecting users to real results. Microsoft says Google removed it from the store after responsible disclosure. The extension was called "Search for perplexity ai" (ID flkebkiofojicogddingbdmcmkpbplcd) and used a look-alike domain, perplexity-ai[.]online, to pass for the real service at perplexity.ai. Microsoft's Defender research team  says the point was to intercept searches and collect data. It found no proof of password theft, but far more access than a search box should ever need. Once installed, the extension sets itself as the browser's default search engine. When you searched, the query went first to perplexity-ai[.]online, where the attacker's server logged it with your browser headers, IP address,...

New research from CrowdStrike has revealed that DeepSeek's artificial intelligence (AI) reasoning model DeepSeek-R1 produces more security vulnerabilities in response to prompts that contain topics deemed politically sensitive by China. "We found that when DeepSeek-R1 receives prompts containing topics the Chinese Communist Party (CCP) likely considers politically sensitive, the likelihood of it producing code with severe security vulnerabilities increases by up to 50%," the cybersecurity company said . The Chinese AI company previously attracted national security concerns, leading to a ban in many countries. Its open-source DeepSeek-R1 model was also found to censor topics considered sensitive by the Chinese government, refusing to answer questions about the Great Firewall of China or the political status of Taiwan, among others. In a statement released earlier this month, Taiwan's National Security Bureau warned citizens to be vigilant when using Chinese-m...

Security researchers at Forever Security have shown that one ordinary browser extension could take control of the AI assistants built into five Chromium-based products: Gemini Live in Chrome, Perplexity Comet, Microsoft Edge, Opera Neon and the Claude in Chrome extension. Once the extension was installed, it could access each product's built-in AI with a single click. On Comet, Edge, Opera Neon, and Claude in Chrome, it could drive the AI agent to act on behalf of the attacker; on Chrome and Comet, it could read files from the user's computer, and on Chrome, it could also switch on the camera and microphone. The findings are researcher demonstrations, not attacks seen in the wild, and each requires the attacker's extension to be already running in the victim's browser. These products all work the same way. The AI has a "body" inside the browser that can see the screen, open files, use the camera, and take actions, and a "brain" that runs on th...

Cybersecurity researchers have disclosed details of a new attack called CometJacking targeting Perplexity's agentic AI browser Comet by embedding malicious prompts within a seemingly innocuous link to siphon sensitive data, including from connected services, like email and calendar. The sneaky prompt injection attack plays out in the form of a malicious link that, when clicked, triggers the unexpected behavior unbeknownst to the victims. "CometJacking shows how a single, weaponized URL can quietly flip an AI browser from a trusted co-pilot to an insider threat," Michelle Levy, Head of Security Research at LayerX, said in a statement shared with The Hacker News. "This isn't just about stealing data; it's about hijacking the agent that already has the keys. Our research proves that trivial obfuscation can bypass data exfiltration checks and pull email, calendar, and connector data off-box in one click. AI-native browsers need security-by-design for agent...

Agentic web browsers that leverage artificial intelligence (AI) capabilities to autonomously execute actions across multiple websites on behalf of a user could be trained and tricked into falling prey to phishing and scam traps. The attack, at its core, takes advantage of AI browsers' tendency to reason their actions and use it against the model itself to lower their security guardrails, Guardio said in a report shared with The Hacker News ahead of publication. "The AI now operates in real time, inside messy and dynamic pages, while continuously requesting information, making decisions, and narrating its actions along the way. Well, 'narrating' is quite an understatement - It blabbers, and way too much!," security researcher Shaked Chen said. "This is what we call Agentic Blabbering : the AI Browser exposing what it sees, what it believes is happening, what it plans to do next, and what signals it considers suspicious or safe." By intercepting thi...

Convince an AI browser that it is playing a game, and it can hand over your login details. That is the finding behind  BioShocking , a technique from security firm LayerX that tricked six AI browsers and assistants into copying a user's credentials and sending them to an attacker. The targets included OpenAI's ChatGPT Atlas, Perplexity's Comet, and Anthropic's Claude browser extension. An AI browser is one that can act for you, not just read pages. Switch it to agent mode, and it can click, type, and reach into the sites you are already signed into. That access is the whole point, and it is also the problem. The trick works because of how these agents read. The web page and your own instructions arrive as a single stream of text. That lets a malicious page slip in commands dressed up as ordinary content or game rules, and the agent cannot reliably tell the difference. Researchers call this  indirect prompt injection . How the trick works

The newly released OpenAI ChatGPT Atlas web browser has been found to be susceptible to a prompt injection attack where its omnibox can be jailbroken by disguising a malicious prompt as a seemingly harmless URL to visit. "The omnibox (combined address/search bar) interprets input either as a URL to navigate to, or as a natural-language command to the agent," NeuralTrust said in a report published Friday. "We've identified a prompt injection technique that disguises malicious instructions to look like a URL, but that Atlas treats as high-trust 'user intent' text, enabling harmful actions." Last week, OpenAI launched Atlas as a web browser with built-in ChatGPT capabilities to assist users with web page summarization, inline text editing, and agentic functions. In the attack outlined by the artificial intelligence (AI) security company, an attacker can take advantage of the browser's lack of strict boundaries between trusted user input and untru...

Cybersecurity researchers have flagged a new security issue in agentic web browsers like OpenAI ChatGPT Atlas that exposes underlying artificial intelligence (AI) models to context poisoning attacks. In the attack devised by AI security company SPLX, a bad actor can set up websites that serve different content to browsers and AI crawlers run by ChatGPT and Perplexity. The technique has been codenamed AI-targeted cloaking . The approach is a variation of search engine cloaking, which refers to the practice of presenting one version of a web page to users and a different version to search engine crawlers with the end goal of manipulating search rankings. The only difference in this case is that attackers optimize for AI crawlers from various providers by means of a trivial user agent check that leads to content delivery manipulation. "Because these systems rely on direct retrieval, whatever content is served to them becomes ground truth in AI Overviews, summaries, or autonom...

A new agentic browser attack targeting Perplexity's Comet browser that's capable of turning a seemingly innocuous email into a destructive action that wipes a user's entire Google Drive contents, findings from Straiker STAR Labs show. The zero-click Google Drive Wiper technique hinges on connecting the browser to services like Gmail and Google Drive to automate routine tasks by granting them access to read emails, as well as browse files and folders, and perform actions like moving, renaming, or deleting content. For instance, a prompt issued by a benign user might look like this: "Please check my email and complete all my recent organization tasks." This will cause the browser agent to search the inbox for relevant messages and perform the necessary actions. "This behavior reflects excessive agency in LLM-powered assistants where the LLM performs actions that go far beyond the user’s explicit request," security researcher Amanda Rousseau said in ...

Attackers aren’t waiting for patches anymore — they are breaking in before defenses are ready. Trusted security tools are being hijacked to deliver malware. Even after a breach is detected and patched, some attackers stay hidden. This week’s events show a hard truth: it’s not enough to react after an attack. You have to assume that any system you trust today could fail tomorrow. In a world where AI tools can be used against you and ransomware hits faster than ever, real protection means planning for things to go wrong — and still staying in control. Check out this week’s update to find important threat news, helpful webinars, useful tools, and tips you can start using right away. ⚡ Threat of the Week Windows 0-Day Exploited for Ransomware Attacks — A security affecting the Windows Common Log File System (CLFS) was exploited as a zero-day in ransomware attacks aimed at a small number of targets, Microsoft revealed. The flaw, CVE-2025-29824, is a privilege escalation vulnerabilit...

This week’s security news is mostly about weak spots. Browsers, bots, sandboxes, AI systems, and email flows all show the same problem in different ways. Everything looks normal until someone tests a small gap and finds a way through. This is not one big break. It is small permissions, weak checks, open systems, and normal tools doing things they were allowed to do. That same pattern runs through the stories below.

Cybersecurity researchers have flagged a "coordinated malware campaign" on the JetBrains Marketplace that has published no less than 15 malicious plugins capable of exfiltrating artificial intelligence (AI) provider keys. "Every plugin poses as an AI coding assistant built on DeepSeek and other large language models, offering chat, commit messages, code review, bug finding, and unit tests," Aikido Security researcher Ilyas Makari said . "They function exactly as advertised. However, the AI provider API key you enter gets exfiltrated to a server controlled by the attacker." The activity is said to have been ongoing since the end of October 2025, with new plugins released as recently as June 10, 2026. Two of the plugins, CodeGPT AI Assistant and DeepSeek AI Assist, have more than 25,000 downloads each, although it's not clear if the counts are authentic or if they have been inflated to fake their popularity. The complete list of plugins is below -...

A Google Chrome extension with a "Featured" badge and six million users has been observed silently gathering every prompt entered by users into artificial intelligence (AI)-powered chatbots like OpenAI ChatGPT, Anthropic Claude, Microsoft Copilot, DeepSeek, Google Gemini, xAI Grok, Meta AI, and Perplexity. The extension in question is Urban VPN Proxy , which has a 4.7 rating on the Google Chrome Web Store. It's advertised as the "best secured Free VPN access to any website, and unblock content." Its developer is a Delaware-based company named Urban Cyber Security Inc . On the Microsoft Edge Add-ons marketplace, it has 1.3 million installations .  Despite claiming that it allows users to "protect your online identity, stay protected, and hide your IP," an update was pushed to users on July 9, 2025, when version 5.5.0 was released with the AI data harvesting enabled by default using hard-coded settings. Specifically, this is achieved by means of a t...

Cybersecurity researchers have disclosed a new set of vulnerabilities impacting OpenAI's ChatGPT artificial intelligence (AI) chatbot that could be exploited by an attacker to steal personal information from users' memories and chat histories without their knowledge. The seven vulnerabilities and attack techniques, according to Tenable, were found in OpenAI's GPT-4o and GPT-5 models. OpenAI has since addressed some of them .  These issues expose the AI system to indirect prompt injection attacks , allowing an attacker to manipulate the expected behavior of a large language model (LLM) and trick it into performing unintended or malicious actions, security researchers Moshe Bernstein and Liv Matan said in a report shared with The Hacker News. The identified shortcomings are listed below - Indirect prompt injection vulnerability via trusted sites in Browsing Context, which involves asking ChatGPT to summarize the contents of web pages with malicious instructions added...

Cybersecurity researchers have disclosed details of a new attack method dubbed Reprompt that could allow bad actors to exfiltrate sensitive data from artificial intelligence (AI) chatbots like Microsoft Copilot in a single click, while bypassing enterprise security controls entirely. "Only a single click on a legitimate Microsoft link is required to compromise victims," Varonis security researcher Dolev Taler said in a report published Wednesday. "No plugins, no user interaction with Copilot." "The attacker maintains control even when the Copilot chat is closed, allowing the victim's session to be silently exfiltrated with no interaction beyond that first click." Following responsible disclosure, Microsoft has addressed the security issue. The attack does not affect enterprise customers using Microsoft 365 Copilot. At a high level, Reprompt employs three techniques to achieve a data‑exfiltration chain - Using the "q" URL parameter in...

According to the new Browser Security Report 2025 , security leaders are discovering that most identity, SaaS, and AI-related risks converge in a single place, the user’s browser. Yet traditional controls like DLP, EDR, and SSE still operate one layer too low. What’s emerging isn’t just a blindspot. It’s a parallel threat surface: unmanaged extensions acting like supply chain implants, GenAI tools accessed through personal accounts, sensitive data copy/pasted directly into prompt fields, and sessions that bypass SSO altogether. This article unpacks the key findings from the report and what they reveal about the shifting locus of control in enterprise security. GenAI Is Now the Top Data Exfiltration Channel The rise of GenAI in enterprise workflows has created a massive governance gap. Nearly half of employees use GenAI tools, but most do so through unmanaged accounts, outside of IT visibility. Key stats from the report: 77% of employees paste data into GenAI prompts 82% of...

A severe vulnerability disclosed in Chromium's Blink rendering engine can be exploited to crash many Chromium-based browsers within a few seconds. Security researcher Jose Pino, who disclosed details of the flaw, has codenamed it Brash . "It allows any Chromium browser to collapse in 15-60 seconds by exploiting an architectural flaw in how certain DOM operations are managed," Pino said in a technical breakdown of the shortcoming. At its core, Brash stems from the lack of rate limiting on " document.title " API updates, which, in turn, allows for bombarding millions of [document object model] mutations per second, causing the web browser to crash, as well as degrade system performance as a result of devoting CPU resources to this process. The attack plays out in three steps - Hash generation or preparation phase, where the attacker preloads into memory 100 unique hexadecimal strings of 512 characters that act as a seed for the browser tab title changes ...

Cybersecurity researchers have disclosed a zero-click flaw in OpenAI ChatGPT's Deep Research agent that could allow an attacker to leak sensitive Gmail inbox data with a single crafted email without any user action. The new class of attack has been codenamed ShadowLeak by Radware. Following responsible disclosure on June 18, 2025, the issue was addressed by OpenAI in early August. "The attack utilizes an indirect prompt injection that can be hidden in email HTML (tiny fonts, white-on-white text, layout tricks) so the user never notices the commands, but the agent still reads and obeys them," security researchers Zvika Babo, Gabi Nakibly, and Maor Uziel said . "Unlike prior research that relied on client-side image rendering to trigger the leak, this attack leaks data directly from OpenAI's cloud infrastructure, making it invisible to local or enterprise defenses." Launched by OpenAI in February 2025, Deep Research is an agentic capability built into ...

Cybersecurity researchers have demonstrated a new prompt injection technique called PromptFix that tricks a generative artificial intelligence (GenAI) model into carrying out unintended actions by embedding the malicious instruction inside a fake CAPTCHA check on a web page. Described by Guardio Labs an "AI-era take on the ClickFix scam," the attack technique demonstrates how AI-driven browsers, such as Perplexity's Comet , that promise to automate mundane tasks like shopping for items online or handling emails on behalf of users can be deceived into interacting with phishing landing pages or fraudulent lookalike storefronts without the human user's knowledge or intervention. "With PromptFix, the approach is different: We don't try to glitch the model into obedience," Guardio researchers Nati Tal and Shaked Chen said . "Instead, we mislead it using techniques borrowed from the human social engineering playbook – appealing directly to its core d...

Cyber threats last week showed how attackers no longer need big hacks to cause big damage. They’re going after the everyday tools we trust most — firewalls, browser add-ons, and even smart TVs — turning small cracks into serious breaches. The real danger now isn’t just one major attack, but hundreds of quiet ones using the software and devices already inside our networks. Each trusted system can become an entry point if it’s left unpatched or overlooked. Here’s a clear look at the week’s biggest risks, from exploited network flaws to new global campaigns and fast-moving vulnerabilities. ⚡ Threat of the Week Flaws in Multiple Network Security Products Come Under Attack — Over the past week, Fortinet , SonicWall , Cisco , and WatchGuard said vulnerabilities in their products have been exploited by threat actors in real-world attacks. Cisco said attacks exploiting CVE-2025-20393, a critical flaw in AsyncOS, have been abused by a China-nexus advanced persistent threat (APT) actor cod...