⚡ Bluetooth car exploits. macOS backdoors. Ransomware returns. Arrests hit top cyber gangs.

This week, trusted tools turned toxic—and no one’s perimeter is safe.

Get the full intel in our latest recap →
https://thehackernews.com/2025/07/weekly-recap-scattered-spider-arrests.html

🚨 A fake CAPTCHA is all it takes.

Interlock ransomware is back—now pushing a stealthy PHP RAT via “FileFix,” a spin on ClickFix that hijacks File Explorer.

Targets? Everyone.
Tactics? Evolving fast.

Here’s what they’re hiding behind Cloudflare Tunnels ↓ https://thehackernews.com/2025/07/new-php-based-interlock-rat-variant.html

🚨 39 million secrets leaked on GitHub in 2024 alone.

Not just mistakes—these are entry points for attackers to breach your cloud, CI/CD, and data stores without triggering alerts.

Even a forgotten repo can cost everything.

Here’s how to stop it → https://thehackernews.com/2025/07/the-unusual-suspect-git-repos.html

🚨 The bait? Fake coding assignments.

North Korean hackers pushed 67 new malware-laced npm packages—over 17K downloads already.

They’re now using a stealthier loader called XORIndex to hijack dev machines, steal crypto, and drop Python backdoors.

Read → https://thehackernews.com/2025/07/north-korean-hackers-flood-npm-registry.html

🚨 AsyncRAT didn’t just survive—it multiplied.

Since its 2019 GitHub release, this open-source trojan has spawned a swarm of powerful variants like DCRat and Venom RAT—stealing webcams, logging keystrokes, evading antivirus, and more.

Details here → https://thehackernews.com/2025/07/asyncrats-open-source-code-sparks-surge.html

🚨 Sick of chasing meaningless vuln alerts?

Function-level runtime reachability shows only the code that's actually executing—no fluff, no false positives.

It’s like turning off the noise and seeing just the real threats.

Here’s why it changes everything ↓ https://thehackernews.com/expert-insights/2025/07/everything-to-know-about-runtime.html

🚨 A new backdoor dubbed HazyBeacon is hitting Southeast Asian governments—hiding in plain sight using AWS Lambda.

It exfiltrates trade secrets via Google Drive and Dropbox.

State-backed. Stealthy. Still active.

Here’s what you need to know ↓ https://thehackernews.com/2025/07/state-backed-hazybeacon-malware-uses.html

LIVE WEBINAR: Building Your Secure AI Roadmap 🔒

Join this practical session to learn how leading teams are integrating security into every phase of the AI lifecycle.

Expect field insights, experts' best practices, and live answers to your toughest questions on AI visibility, red-teaming, runtime protection, and governance.

🗓️ July 23rd, 10 AM ET
🔗 Register here: https://thn.news/secure-ai-roadmap

🚨 AI agents now outnumber humans in the cloud — and they're logging in with invisible keys.

They act fast, never question commands, and often hold overprivileged access no one monitors.

One forgotten token = full-blown breach.

Here’s how to fix it before it breaks you → https://thehackernews.com/2025/07/securing-agentic-ai-how-to-protect.html

🚨 New ransomware gang "GLOBAL GROUP" just hit 17 victims across 4 continents — in weeks.

It’s a rebrand of BlackLock, now upgraded with AI chatbots and 85% payouts to lure affiliates.

They’re buying access to law firms and critical networks.

Full story → https://thehackernews.com/2025/07/newly-emerged-global-group-raas-expands.html

🚨 A single DDoS attack just peaked at 7.3 Tbps—in 45 seconds.

Cloudflare says hyper-volumetric attacks are exploding, with ransom threats up 68%.

Botnets like DemonBot are turning unsecured IoT into weapons.

Here’s what’s happening → https://thehackernews.com/2025/07/hyper-volumetric-ddos-attacks-reach.html

🛑 Google’s AI just stopped a cyberattack before it even began.

An LLM agent named Big Sleep discovered a critical SQLite flaw (CVE-2025-6965) that threat actors were likely about to exploit—making it the first known case of an AI foiling a live exploit attempt.

Details on how it worked → https://thehackernews.com/2025/07/google-ai-big-sleep-stops-exploitation.html

⚠️ WARNING - Google Chrome Zero-Day Alert!

A new exploit is hitting users right now — CVE-2025-6558 lets attackers break out of the browser sandbox.

Update Chrome ASAP → https://thehackernews.com/2025/07/urgent-google-releases-critical-chrome.html

(Edge, Brave, Opera users: you're not safe either.)

🚨 Two Android threats just collided — and it’s getting wild.

Hackers are using evil twin apps with the same package name as legit Play Store apps to deliver Konfety malware — now upgraded with fake encryption, APK crashes, dynamic payloads, and hidden installs.

At the same time, TapTrap lets apps hijack your screen taps to grant malware dangerous permissions — no user awareness needed.

This isn’t theoretical. Android 16 is still vulnerable.

Details here → https://thehackernews.com/2025/07/new-konfety-malware-variant-evades.html

🚨 M&A chaos isn't just financial—it’s identity chaos too.

Mergers can spawn 25+ siloed systems, untracked privileged accounts, and dangerous role creep. One wrong move? Audit failures or full-blown breaches.

Here’s how to restore order before risk takes over ↓ https://thehackernews.com/expert-insights/2025/07/identity-challenges-in-mergers-and.html

⚡ Deepfake CEOs. Fake job offers. Cloned websites.

AI-powered scams are fooling even trained teams—because they look real. These aren’t phishing emails. They’re full-scale impersonation attacks.

See how to stop them in real time.

Watch the webinar → https://thehackernews.com/2025/07/deepfakes-fake-recruiters-cloned-cfos.html

🚨 One flaw. Total takeover.

A critical bug in Windows Server 2025 lets attackers generate any dMSA/gMSA password—across the forest.

It bypasses Credential Guard and can persist for years.

Details here → https://thehackernews.com/2025/07/critical-golden-dmsa-attack-in-windows.html

🚨 Your AI agent isn’t just helping—it has full access to code, payroll, email, and more.

If one device gets hacked, it can leak everything—fast.

The real risk? It’s not the AI. It’s who (or what) can talk to it.

Learn how to lock it down → https://thehackernews.com/2025/07/ai-agents-act-like-employees-with-root.html

ISPs are high-value DDoS targets—and need tailored defenses.

We tested the DDoS resilience of a leading mobile operator by simulating six attacks that bypassed their CDN and hit the origin server directly.

While network-layer attacks were blocked, two app-layer attacks—a GET Flood and POST Flood—slipped through.

Read the full case study and our hardening recommendations → https://thn.news/african-telecom-ddos

🚨 Hackers are exploiting fully patched SonicWall SMA 100 firewalls using stolen OTP seeds and a stealth backdoor.

Google says the malware, OVERSTEP, rewrites boot scripts, hides logs, and steals credentials—persisting even after reboots.

Full story → https://thehackernews.com/2025/07/unc6148-backdoors-fully-patched.html