A flaw in @Adobe's Acrobat Chrome extension (CVE-2026-48294) could let attacker-controlled websites access @WhatsApp Web data after a single visit.
@GuardioSecurity Labs says the issue affected an extension installed on roughly 329 million browsers. Adobe has released a patched version.
Guardio retweeted
On the menu in Vietnam: pho, banh mi, dumplings... and 30,000 hijacked Facebook business accounts 🥟
Meet "AccountDumpling": phishing sent straight through Google's own infrastructure. Yum.
Looks like it's a serious business over there. Steal the account. Resell it. Fabricate fake identities. Sell "recovery" right back to the victim for a fee and even more shenanigans.
Brilliant deep-dive research by our own @shaked__chen! Learn how they abused Google to send phishing emails, so many methods and tricks they used along the way, and the mistakes they made that helped us catch them red-handed and reach victims in time to save what could still be saved.
(link to full research in reply...)
Guardio retweeted
NoMetaWhat. Ever got a phishing email from Meta that actually came from Facebook ?! Oh No.. Attackers abuse Facebook Business Manager to send real emails from [email protected], then stuff the invite with their own sender name, urgency bait, and a clickable phishing link!
DKIM, SPF, DMARC all pass. Of course they do.
👉 more...
Guardio retweeted
Didn’t realize a church in Pennsylvania started offering macOS storage fix services?!
The ClickFix gang is back with GoogleFix, and they’ve figured out how to make Google’s own ad platform do their dirty work.
Guardio Labs' full report now available with a technical deep dive, and tons of in-the-wild samples here 👇
Guardio retweeted
#GoogleFix is on the roll! Searching for "Claude Code" on Google? Don’t click the top result! Again! 🤦♂️ Attackers now use a sponsored Squarespace link that perfectly clones the official Claude Code docs, except the install commands are altered to drop Infostealers instead of Claude Code Agent! This time both Windows and MacOS are targeted! 👉 more...
Guardio retweeted
Don't fall for this Amazon recall text scam. It's designed to hack your account so the scammer can take over your account. #MoneyMatters #DontClicktheLink @FOX59
fox59.com/news/watch-out-for…
Guardio retweeted
Grammarly now joins the endless stream of reputable platforms exploited in the #GoogleFix campaign.
Attackers seed fake Apple support guides, now even on Grammarly, @WriteSonic , @evernote , Claude, then push these via Hijacked @GoogleAds accounts. One careless terminal command leads to AMOS infostealer on your Mac.
The scale and persistence here is remarkable... Stay vigilant! --> Get @GuardioSecurity 😎
Guardio retweeted
1/3 Here’s how malvertising campaigns weaponize AI chat tools - propagating malicious tech support instructions as if the AI Chat actually suggested them!
It starts with simply asking the AI to always reply with printing back a predefined instruction set to the screen. The AI, as always, complies: "Your wish is my (malicious) command!" 🤖
Guardio retweeted
Security startup Guardio nabs $80M from ION Crossover Partners techcrunch.com/2025/11/19/se…
Guardio retweeted
🛡️ @GuardioSecurity Head of Research, Nati Tal, explains how AI is reshaping cybercrime - and why “technology protects you in the moment, but knowledge protects you for life.”
He warns: “With AI browsers, AI itself will be falling for scams while browsing on our behalf.”
What’s your take on AI-driven fraud?
#AI #Phishing #AIFraud #Guardio #Cybersecurity #SocialEngineering
Guardio retweeted
At Lovable, we’re not just making it easier to create software. We’re making sure those creations make the web better for everyone.
Every site published on Lovable is now scanned by Guardio to prevent phishing, scams, and impersonation.
Every site published on Lovable is now scanned by @GuardioSecurity for phishing, scams, and impersonation.
As AI-powered tools democratize digital creation, we ensure these new tools are not used to cause harm.
This is what responsible platform-building looks like.
Guardio retweeted
Every site published on Lovable is now scanned by @GuardioSecurity for phishing, scams, and impersonation.
As AI-powered tools democratize digital creation, we ensure these new tools are not used to cause harm.
This is what responsible platform-building looks like.
Guardio retweeted
Grok says @elonmusk is the President?! AI races to become the fact-checker-single-source-of-truth. @Wikipedia is dead!? Long live the new @Grokipedia!?
We are reaching the point where funny prompt injections might just escalate to truth-manipulating Prompt Inception!
👇 Read the complete research (link in reply)
Guardio retweeted
Phishing has never been more deceptive.
Look at how attackers target high-value Facebook Business accounts managing serious budgets.
The trick? A link to your "copyright-infringing" ad, followed by a fake Meta Captcha and a "browser-in-the-browser" login that looks so real! (1/3)
Guardio retweeted
If you decide to use ChatGPT Atlas (@OpenAI's new browser), make double sure you don't reuse your password for OpenAI AND have strong MFA on.
If that reused password shows up in a breach & can be used against your AI browser, you're gonna have a much bigger breach on your hands.
OpenAI@OpenAI21 Oct 2025🇺🇸
United StatesConnected via WebAccount-level information, not a live location or per-post device.
Meet our new browser—ChatGPT Atlas.
Available today on macOS: chatgpt.com/atlas
Guardio retweeted
AWS outage crashes Amazon, PrimeVideo, Fortnite, Perplexity and more - @mayank_jee
bleepingcomputer.com/news/te…
bleepingcomputer.com/news/te…
Guardio retweeted
0G was hacked, but hacks happen every time in the crypto space.
So, how do you protect yourself from falling victim to these scam links?
Here are a few things you can do to protect yourself👇🏽
#1. Verify Links With ChatGPT
3 days ago, I was almost hacked by someone who pretended to want to offer me a job. But how did I protect myself?
I copied the link and dropped it into ChatGPT, then asked it to check if it was a scam link. ChatGPT ran the link, told me it was a scam link, and explained why and how exactly I'd have been hacked if I had used the link.
#2. Security Plugins
There are actually several security plugins you can use to detect scam links before you mistakenly click them.
Examples include @GuardioSecurity and @realScamSniffer. Some wallets like OKX, Rabby and MetaMask also have in-built security systems for detecting these scam links
I have this particular plugin installed
#3. Always Verify Links
Before you click any link, make sure you're getting it directly from official sources, like the project's website or X handle.
Even if it's a site you're familiar with, there's always the risk of domain names expiring and getting taken over by hackers (it has even happened to Google before!)
#4. Revoke Permissions
Someone just lost $90k+ due to a phishing approval signed 499 days ago!
xitter.cf/realScamSniffer/status…
It tells you that even if a link or site is genuine when you're interacting with it, there's no way to predict what could happen 5 or 6 months down the line.
Use tools like @RevokeCash and @realScamSniffer to review and revoke old approvals
Scam Sniffer@realScamSniffer5 Oct 2025🇯🇵!
JapanConnected via China App StoreLocation may be affected by a proxy or VPN.Account-level information, not a live location or per-post device.
🚨 ALERT: Another victim lost $90,691 due to a phishing approval signed 499 days ago.
🔐 REMINDER: Regularly review and revoke old approvals - your wallet security matters! 💰
Guardio retweeted

PCMag@PCMag7 Sep 2025
🇺🇸
United StatesConnected via WebAccount-level information, not a live location or per-post device.
Cybercriminals run video ads on X containing adult content to grab scrollers’ attention, then trick Grok into amplifying malicious links in the response below. pcmag.com/news/using-grok-be…
Guardio retweeted
Scammers Are Using Grok to Spread Malicious Links on X: bit.ly/47wVfHO by Nate Nelson
Guardio retweeted
Cybercriminals are exploiting a loophole with X’s AI assistant Grok.
They trick it into sharing malicious links in promoted posts—links that normally should be blocked—reaching millions of users.
Researchers call it “Grokking.”
Here’s how it works ↓ thehackernews.com/2025/09/cy…
























