NEURALSHIELD
AI System Protection Engineer
Professional Profile
Defends AI systems from attacks such as poisoning and inversion.
Key Skills
AI security; Data-poisoning mitigation; Model-inversion defense; Dataset analysis; AI-pipeline hardening
Major Achievements
Stopped an attack that would have manipulated millions of automated decisions.
Articles by NEURALSHIELD
The MFA Illusion: Why Stolen Microsoft 365 Sessions Are the New Prize
A reported phishing operation built around Evilginx2-style relays shows how attackers can win after the login succeeds by reusing the session, not cracking the password.
Microsoft Teams Draws a Curtain Over QR Code Lures in External Chats
A new reveal-before-view step adds friction to QR-code phishing in Teams conversations that cross organizational boundaries.
Microsoft Turns QR Codes into a Suspicious Object in Teams
A planned Teams control would hide QR codes from external senders, adding friction to a familiar phishing path that often depends on speed, habit, and misplaced trust.
Invisible Characters, Visible Risk: The Phishing Trick That Slips Past Simple Filters
A finance-themed phishing wave used hidden Unicode tag characters to split suspicious words at the processing layer, exposing how text security can fail when it only reads what the eye can see.
When the QR Code Is Not an Image: The New Blind Spot in Email Security
A phishing lure can now be built from HTML itself, creating a visual QR code that may slip past scanners designed to inspect attachments, bitmaps, or OCR output.
Inside the Teams Trap: How a Help-Desk Persona Became an Attack Vector
A reported Spring Ring campaign shows how external Microsoft Teams accounts, voice phishing, and Windows authentication abuse can turn a trusted collaboration channel into an access path.
When a QR Code Pretends to Be Plain Text, Email Defenses Can Miss the Trap
Text-rendered QR codes are giving quishing campaigns a quieter delivery method, pushing the user from email to a mobile phishing page.
The Intrusion Chain That Hid in Plain Sight
A fake IT-support lure, a malicious MSI, and trusted-looking cloud infrastructure show how modern intrusions are being broken into quieter pieces to dodge detection.
Teams Became the New Phishing Front Door, and IT Impersonators Walked Right Through It
A campaign dubbed Spring Ring shows how trusted collaboration tools can be turned into social-engineering lanes, with fake help desk identities and live voice calls replacing the old email lure.
Microsoft Teams Becomes the Lure in a Multi-Stage IT-Help Scam
A Spring Ring campaign turned a familiar collaboration tool into a trust trap, pairing vishing, RMM abuse, and NTLM relay techniques without any public evidence of a Teams software flaw.
Android 17 Turns the Spotlight on the Hidden Layer of Mobile Attacks
Google’s latest Android network-security push is aimed at the quiet tricks attackers use before encryption ever begins: radio downgrades, Wi-Fi discovery, and shaky certificate trust.
Android 17 Tightens the Screws on 2G Downgrade Risk
Google’s latest Android network controls target legacy cellular fallback, SMS blaster abuse, and a less obvious privacy problem: what encrypted sessions can still reveal.
When a Fake Verification Box Becomes a Network Entry Point
A deceptive CAPTCHA page is being used to push victims into running a PowerShell command that can create a reverse tunnel back into internal systems.
The Router Becomes the Gatekeeper in Home-Network DNS Defense
Pointing a home router at Cloudflare’s filtering resolvers can add a quiet layer against malicious domains, but it works only as part of a broader defense-in-depth setup.
The Phish That Snared Itself: A JavaScript Loop Broke the Steal
A page built to dodge security tools appears to have collapsed under its own browser logic, turning a credential theft attempt into a dead end.
The Scam That Does Not Break Apple - It Tries to Outtalk It
An AI-enabled phishing service is described as pushing victims through account, passcode, and 2FA prompts to help remove Activation Lock through social engineering, not cryptographic defeat.
The iPhone Lock That Crime Wants to Pick With a Voice
A reported phishing service aimed at Apple users shows how stolen-device monetization is shifting from hardware tampering to identity theft, code theft, and social engineering.
The Session Was the Prize: How AitM Kits Turn Microsoft 365 Login Pages Into Profit
A subscription phishing toolkit marketed for Microsoft 365 shows how cybercrime is moving from password capture to session hijacking, where the real value sits in the token, not the login.
When a Fake Signature Trail Becomes a Live Microsoft 365 Trap
NovaCookies turns trusted document notifications into a relay point for identity theft, showing how session capture can matter more than password theft.
When a Package Mirror Becomes a Phishing Stage
A small cluster of malicious npm packages was used to host fake Cloudflare verification pages, showing how trusted developer infrastructure can be repurposed as a browser-facing lure.