Technology
OpenAI admits AI browsers face unsolvable prompt attacks
NEWYou can now listen to Fox News articles!
Cybercriminals don’t always need malware or exploits to break into systems anymore. Sometimes, they just need the right words in the right place. OpenAI is now openly acknowledging that reality. The company says prompt injection attacks against artificial intelligence (AI)-powered browsers are not a bug that can be fully patched, but a long-term risk that comes with letting AI agents roam the open web. This raises uncomfortable questions about how safe these tools really are, especially as they gain more autonomy and access to your data.
Sign up for my FREE CyberGuy Report
Get my best tech tips, urgent security alerts, and exclusive deals delivered straight to your inbox. Plus, you’ll get instant access to my Ultimate Scam Survival Guide — free when you join my CYBERGUY.COM newsletter.
NEW MALWARE CAN READ YOUR CHATS AND STEAL YOUR MONEY
AI-powered browsers can read and act on web content, which also makes them vulnerable to hidden instructions attackers can slip into pages or documents. (Kurt “CyberGuy” Knutsson)
Why prompt injection isn’t going away
In a recent blog post, OpenAI admitted that prompt injection attacks are unlikely to ever be completely eliminated. Prompt injection works by hiding instructions inside web pages, documents or emails in ways that humans don’t notice, but AI agents do. Once the AI reads that content, it can be tricked into following malicious instructions.
OpenAI compared this problem to scams and social engineering. You can reduce them, but you can’t make them disappear. The company also acknowledged that “agent mode” in its ChatGPT Atlas browser increases risk because it expands the attack surface. The more an AI can do on your behalf, the more damage it can cause when something goes wrong.
OpenAI launched the ChatGPT Atlas browser in October, and security researchers immediately started testing its limits. Within hours, demos appeared showing that a few carefully placed words inside a Google Doc could influence how the browser behaved. That same day, Brave published its own warning, explaining that indirect prompt injection is a structural problem for AI-powered browsers, including tools like Perplexity’s Comet.
This isn’t just OpenAI’s problem. Earlier this month, the National Cyber Security Centre in the U.K. warned that prompt injection attacks against generative AI systems may never be fully mitigated.
FAKE AI CHAT RESULTS ARE SPREADING DANGEROUS MAC MALWARE
Prompt injection attacks exploit trust at scale, allowing malicious instructions to influence what an AI agent does without the user ever seeing it. (Kurt “CyberGuy” Knutsson)
The risk trade-off with AI browsers
OpenAI says it views prompt injection as a long-term security challenge that requires constant pressure, not a one-time fix. Its approach relies on faster patch cycles, continuous testing, and layered defenses. That puts it broadly in line with rivals like Anthropic and Google, which have both argued that agentic systems need architectural controls and ongoing stress testing.
Where OpenAI is taking a different approach is with something it calls an “LLM-based automated attacker.” In simple terms, OpenAI trained an AI to act like a hacker. Using reinforcement learning, this attacker bot looks for ways to sneak malicious instructions into an AI agent’s workflow.
The bot runs attacks in simulation first. It predicts how the target AI would reason, what steps it would take and where it might fail. Based on that feedback, it refines the attack and tries again. Because this system has insight into the AI’s internal decision-making, OpenAI believes it can surface weaknesses faster than real-world attackers.
Even with these defenses, AI browsers aren’t safe. They combine two things attackers love: autonomy and access. Unlike regular browsers, they don’t just display information, but also read emails, scan documents, click links and take actions on your behalf. That means a single malicious prompt hidden in a webpage, document or message can influence what the AI does without you ever seeing it. Even when safeguards are in place, these agents operate by trusting content at scale, and that trust can be manipulated.
THIRD-PARTY BREACH EXPOSES CHATGPT ACCOUNT DETAILS
As AI browsers gain more autonomy and access to personal data, limiting permissions and keeping human confirmation in the loop becomes critical for safety. (Kurt “CyberGuy” Knutsson)
7 steps you can take to reduce risk with AI browsers
You may not be able to eliminate prompt injection attacks, but you can significantly limit their impact by changing how you use AI tools.
1) Limit what the AI browser can access
Only give an AI browser access to what it absolutely needs. Avoid connecting your primary email account, cloud storage or payment methods unless there’s a clear reason. The more data an AI can see, the more valuable it becomes to attackers. Limiting access reduces the blast radius if something goes wrong.
2) Require confirmation for every sensitive action
Never allow an AI browser to send emails, make purchases or modify account settings without asking you first. Confirmation breaks long attack chains and gives you a moment to spot suspicious behavior. Many prompt injection attacks rely on the AI acting quietly in the background without user review.
3) Use a password manager for all accounts
A password manager ensures every account has a unique, strong password. If an AI browser or malicious page leaks one credential, attackers can’t reuse it elsewhere. Many password managers also refuse to autofill on unfamiliar or suspicious sites, which can alert you that something isn’t right before you manually enter anything.
Next, see if your email has been exposed in past breaches. Our #1 password manager (see Cyberguy.com) pick includes a built-in breach scanner that checks whether your email address or passwords have appeared in known leaks. If you discover a match, immediately change any reused passwords and secure those accounts with new, unique credentials.
Check out the best expert-reviewed password managers of 2025 at Cyberguy.com
4) Run strong antivirus software on your device
Even if an attack starts inside the browser, antivirus software can still detect suspicious scripts, unauthorized system changes or malicious network activity. Strong antivirus software focuses on behavior, not just files, which is critical when dealing with AI-driven or script-based attacks.
The best way to safeguard yourself from malicious links that install malware, potentially accessing your private information, is to have strong antivirus software installed on all your devices. This protection can also alert you to phishing emails and ransomware scams, keeping your personal information and digital assets safe.
Get my picks for the best 2025 antivirus protection winners for your Windows, Mac, Android & iOS devices at Cyberguy.com
5) Avoid broad or open-ended instructions
Telling an AI browser to “handle whatever is needed” gives attackers room to manipulate it through hidden prompts. Be specific about what the AI is allowed to do and what it should never do. Narrow instructions make it harder for malicious content to influence the agent.
6) Be careful with AI summaries and automated scans
When an AI browser scans emails, documents or web pages for you, remember that hidden instructions can live inside that content. Treat AI-generated actions as drafts or suggestions, not final decisions. Review anything the AI plans to act on before approving it.
7) Keep your browser, AI tools and operating system updated
Security fixes for AI browsers evolve quickly as new attack techniques emerge. Delaying updates leaves known weaknesses open longer than necessary. Turning on automatic updates ensures you get protection as soon as they’re available, even if you miss the announcement.
CLICK HERE TO DOWNLOAD THE FOX NEWS APP
Kurt’s key takeaway
There’s been a meteoric rise in AI browsers. We’re now seeing them from major tech companies, including OpenAI’s Atlas, The Browser Company’s Dia, and Perplexity’s Comet. Even existing browsers like Chrome and Edge are pushing hard to add AI and agentic features into their current infrastructure. While these browsers can be useful, the technology is still early. It’s best not to fall for the hype and to wait for it to mature.
Do you think AI browsers are worth the risk today, or are they moving faster than security can keep up? Let us know by writing to us at Cyberguy.com
Sign up for my FREE CyberGuy Report
Get my best tech tips, urgent security alerts, and exclusive deals delivered straight to your inbox. Plus, you’ll get instant access to my Ultimate Scam Survival Guide — free when you join my CYBERGUY.COM newsletter.
Copyright 2025 CyberGuy.com. All rights reserved.
Technology
Amazon’s smart shopping cart for Whole Foods gets bigger, lighter, and adds tap-to-pay
Amazon is launching a revamped version of its smart shopping cart, which it plans to bring to dozens of Whole Foods locations by the end of this year, according to an announcement on Wednesday. The new Dash Cart features a “more responsive” item scanner that’s now located next to the built-in display, along with a new NFC reader that lets you tap to pay with your credit card or phone.
Amazon’s previous Dash Cart design put scanners beneath and in front of the handle, potentially making them harder to spot. It also only let you pay with the credit card attached to your Amazon account.
With the upgraded Dash Cart, you’ll find a new scale alongside the cart’s handle, which Amazon says “works in tandem with on-cart cameras, weight sensors, and deep learning models to ensure accurate pricing for every item.” The upgraded Dash Cart eliminates the large sensors facing inside the cart as well, offering a 40 percent larger capacity and a 25 percent lighter weight.
The Dash Cart shows an interactive map of the store on its display, similar to Instacart’s smart Caper Cart. You can sync your shopping list created with Alexa, too, and see how much you’re spending as you add more items to your cart. The cart uses built-in sensors and computer vision to detect when you’ve removed an item, allowing it to automatically update your total. When you’re done shopping, you can skip the checkout line and leave the store in a designated Dash Cart lane.
Amazon is launching its new Dash Cart as the company shakes up its grocery business, which has tied Whole Foods more closely to the Amazon brand. The company has already brought its new Dash Cart to three Whole Foods stores in McKinney, Texas; Reston, Virginia; and Westford, Massachusetts, along with two Amazon Fresh stores.
Technology
Fake error popups are spreading malware fast
NEWYou can now listen to Fox News articles!
A dangerous cybercrime tool has surfaced in underground forums, making it far easier for attackers to spread malware.
Instead of relying on hidden downloads, this tool pushes fake error messages that pressure you into fixing problems that never existed. Security researchers say this method is spreading quickly because it feels legitimate. The page looks broken. The warning feels urgent. The fix sounds simple.
That combination is proving alarmingly effective for cybercriminals.
Sign up for my FREE CyberGuy Report
Get my best tech tips, urgent security alerts and exclusive deals delivered straight to your inbox. Plus, you’ll get instant access to my Ultimate Scam Survival Guide – free when you join my CYBERGUY.COM newsletter.
How fake error malware attacks actually work
These attacks begin with a compromised website. When a visitor lands on the page, something looks wrong right away. Text appears broken. Fonts look scrambled. Visual elements seem corrupted. A pop-up then appears claiming the issue can be fixed with a browser update or a missing system font. A button offers to repair the problem instantly.
Clicking that button copies a command to the clipboard and displays instructions to paste it into PowerShell or a system terminal. That single step launches the infection.
MALICIOUS CHROME EXTENSIONS CAUGHT STEALING SENSITIVE DATA
Fake error popups make a website look broken by scrambling text or fonts to create urgency and panic. (Jens Büttner/picture alliance via Getty Images)
Why this new tool changes the threat landscape
The tool behind these attacks is called ErrTraffic. It automates the entire process and removes the technical barriers that once limited cybercrime operations. For about $800, attackers get a full package with a control panel and scripted payload delivery. Analysts at the Hudson Rock Threat Intelligence Team identified the tool after tracking its promotion on Russian-language forums in early December 2025.
ErrTraffic works through a simple JavaScript injection. A single line of code connects a hacked site to the attacker’s dashboard. From there, everything adapts automatically. The script detects the operating system and browser. It then displays a customized fake error message in the correct language. The attack works across Windows, Android, macOS and Linux.
MOST PARKED DOMAINS NOW PUSH SCAMS AND MALWARE
The popups often claim a browser update or missing system font is needed to fix the problem. (Daniel Acker/Bloomberg via Getty Images)
Why security software struggles to stop it
Traditional malware defenses look for suspicious downloads or unauthorized installations. ErrTraffic avoids both. Browsers see normal text copying. Security tools see a legitimate system utility being opened manually. Nothing appears out of place. That design allows the attack to slip through protections that would normally stop malware in its tracks.
The success rate is deeply concerning
Data pulled from active ErrTraffic campaigns shows conversion rates approaching 60%. That means more than half of the visitors who see the fake error message follow the instructions and install malware. Once active, the tool can deliver infostealers like Lumma or Vidar on Windows devices. Android targets often receive banking trojans instead. The control panel even includes geographic filtering, with built-in blocks for Russia and neighboring regions to avoid drawing attention from local authorities.
What happens after infection?
Once malware is installed, credentials and session data are stolen. Those compromised logins are then used to breach additional websites. Each newly hacked site becomes another delivery vehicle for the same attack. That cycle allows the campaign to grow without direct involvement from the original operator.
FAKE WINDOWS UPDATE PUSHES MALWARE IN NEW CLICKFIX ATTACK
Following the on-screen instructions can quietly trigger malware that steals passwords and personal data. (Kurt Knutsson)
Ways to stay safe from fake error malware
A few smart habits can significantly reduce risk when facing fake error pop-ups and browser-based traps.
1) Never run commands suggested by a website
Legitimate websites never ask you to copy and paste commands into PowerShell or a system terminal. Fake error malware relies on convincing messages that pressure you into doing exactly that. If a page instructs you to run code to fix a problem, close it immediately.
2) Close pages that claim your system is corrupted
Fake error campaigns often use broken text, scrambled fonts or warnings about missing files to grab attention. As a result, these visuals create urgency and trigger fear. In reality, a real system problem never announces itself through a random website, so close the page right away.
3) Install updates only through official system settings
Real browser and operating system updates come from built-in update tools, not pop-ups on websites. If an update is needed, your device will notify you directly through system settings or trusted app stores.
4) Install strong antivirus software on every device
Strong antivirus software can help block malicious scripts, detect infostealers and stop suspicious behavior before damage spreads. This is especially important since fake error malware targets Windows, Android, macOS and Linux systems.
The best way to safeguard yourself from malicious links that install malware, potentially accessing your private information, is to have strong antivirus software installed on all your devices. This protection can also alert you to phishing emails and ransomware scams, keeping your personal information and digital assets safe.
Get my picks for the best 2025 antivirus protection winners for your Windows, Mac, Android and iOS devices at Cyberguy.com.
5) Use a data removal service to reduce exposure
Stolen credentials fuel the spread of fake error malware. Removing personal information from data broker sites can reduce the impact if login details are compromised and limit how far an attack can spread.
While no service can guarantee the complete removal of your data from the internet, a data removal service is really a smart choice. They aren’t cheap, and neither is your privacy. These services do all the work for you by actively monitoring and systematically erasing your personal information from hundreds of websites. It’s what gives me peace of mind and has proven to be the most effective way to erase your personal data from the internet. By limiting the information available, you reduce the risk of scammers cross-referencing data from breaches with information they might find on the dark web, making it harder for them to target you.
Check out my top picks for data removal services and get a free scan to find out if your personal information is already out on the web by visiting Cyberguy.com.
Get a free scan to find out if your personal information is already out on the web: Cyberguy.com.
6) Treat font and browser update pop-ups with suspicion
Claims about missing fonts or outdated browsers are a hallmark of these attacks. Modern systems manage fonts automatically, and browsers update themselves. A webpage has no reason to request manual fixes.
If a real update is needed, the operating system will request it directly. A random webpage never should.
Kurt’s key takeaways
Fake error malware works because it plays on a very human reaction. When something on a screen suddenly looks broken, most people want to fix it fast and move on. That split-second decision is exactly what attackers are counting on. Tools like ErrTraffic show how polished these scams have become. The messages look professional. The instructions feel routine. Nothing about the moment screams danger. But behind the scenes, one click can quietly hand over passwords, banking access and personal data. The good news is that slowing down makes a real difference. Closing a suspicious page and trusting built-in system updates can stop these attacks cold. When it comes to pop-ups claiming your device is broken, walking away is often the smartest fix.
Have you ever seen a pop-up or error message that made you stop and wonder if it was real? Tell us what it looked like and how you handled it by writing to us at Cyberguy.com.
CLICK HERE TO DOWNLOAD THE FOX NEWS APP
Sign up for my FREE CyberGuy Report
Get my best tech tips, urgent security alerts and exclusive deals delivered straight to your inbox. Plus, you’ll get instant access to my Ultimate Scam Survival Guide – free when you join my CYBERGUY.COM newsletter.
Copyright 2025 CyberGuy.com. All rights reserved.
Technology
Intel is planning a custom Panther Lake CPU for handheld PCs
Intel announced yesterday that it’s developing an entire “handheld gaming platform” powered by its new Panther Lake chips, and joining an increasingly competitive field. Qualcomm is hinting about potential Windows gaming handhelds showing up at the Game Developers Conference in March, and AMD’s new Strix Halo chips could lead to more powerful handhelds.
According to IGN and TechCrunch, sources say Intel is going to compete by developing a custom Intel Core G3 “variant or variants” just for handhelds that could outperform the Arc B390 GPU on the chips it just announced. IGN reports that by using the new 18A process, Intel can cut different die slices, and “spec the chips to offer better performance on the GPU where you want it.”
As for concrete details about the gaming platform, we’re going to have to wait. According to Intel’s Dan Rogers yesterday, the company will have “more news to share on that from our hardware and software partners later this year.” The Intel-based MSI Claw saw a marked improvement when it jumped to Lunar Lake, and hopefully the new platform keeps up that positive trend.
-
World1 week agoHamas builds new terror regime in Gaza, recruiting teens amid problematic election
-
News1 week agoFor those who help the poor, 2025 goes down as a year of chaos
-
Science1 week agoWe Asked for Environmental Fixes in Your State. You Sent In Thousands.
-
Business1 week agoA tale of two Ralphs — Lauren and the supermarket — shows the reality of a K-shaped economy
-
Detroit, MI4 days ago2 hospitalized after shooting on Lodge Freeway in Detroit
-
Politics1 week agoCommentary: America tried something new in 2025. It’s not going well
-
Politics1 week agoMarjorie Taylor Greene criticizes Trump’s meetings with Zelenskyy, Netanyahu: ‘Can we just do America?’
-
Health1 week agoRecord-breaking flu numbers reported in New York state, sparking warnings from officials