v1.5.0 · Windows x64 · Live Cloud Backend

DIGITAL FORENSICS REDEFINED BY AI

Entity X silently wraps a live browser in a real-time AI monitoring layer — detecting deepfakes, AI-generated content, and misinformation as you browse. No setup. No expertise required.

0%Detection Rate
0Threat Vectors
0AI Models
0msSetup Time
Core Capabilities

WHAT ENTITY X
DEFENDS YOU FROM

8-layer forensic AI stack — image · video · audio · text · articles · legal · alerts · intelligence.

01
IMAGE DEEPFAKES

5-model ensemble: ViT (99.3%), SwinV2, UniversalFakeDetect. Whistleblower rule enforced.

✓ 99.3% accuracy
02
VIDEO DETECTION

Frame-by-frame analysis detects AI-generated, deepfake or manipulated footage in real-time.

✓ Frame analysis
03
AUDIO DETECTION

Detects AI-cloned voices and synthetic audio using spectral analysis and RNN classifiers.

✓ Voice synthesis
04
AI TEXT DETECTION

RoBERTa classifier + statistical burstiness to find AI-written content anywhere.

✓ ChatGPT-detector
05
ARTICLE FORENSICS

Gemini 2.0 Flash analyzes credibility, bias, and source verification in real-time.

✓ Real-time AI
06
LEGAL AI ENGINE

Auto-generates legal complaints and packages evidence. Legal chat via DeepSeek R1.

✓ PDF export
07
REAL-TIME ALERTS

Silent monitoring fires instant alert toasts the moment any threat is detected.

✓ Background monitor
08
THREAT INTELLIGENCE

Domain intelligence, live threat map, case management, community tracking.

✓ Live Watchlist
AI Stack

THE MODELS
BEHIND THE TRUTH

No single model. No single point of failure. A multi-model ensemble.

ViT · deepfake_vs_real
Vision Transformer
99.3%
SwinV2 · ai-detector
Shifted Window Transformer
98.1%
UniversalFakeDetect
CVPR 2023 SOTA
SOTA
ENTITY-X TERMINAL
entity-x scan --image suspect.jpg ViT ready · 5-model init   ViT: 96.4% FAKE SwinV2: 91.7% FAKE CLIP: 79.1% FAKE   Final: 🔴 DEEPFAKE CONFIRMED Trust: -96 pts [CRITICAL]
Trust System

SESSION
INTEGRITY
IN REAL-TIME

Every session starts at 100/100. As you encounter suspicious content, your Trust Score degrades.

trust_score -= fake_prob × 100
clamped 0–100 · per detection
// SESSION SCORE
87
HIGH INTEGRITY
90–100 · High
70–89 · Moderate
50–69 · Caution
0–49 · Critical
Process

HOW IT WORKS

01
INSTALL

Download the .exe installer. No Python. No API keys. Done in seconds.

02
BROWSE

Use the built-in browser. The AI layer runs silently in the background.

03
DETECT

Every media is analyzed by the 5-model ensemble instantly.

04
DEFEND

Generate legal complaints, package evidence automatically.

Free · Open Source · MIT License

PROTECT YOUR
DIGITAL TRUTH

No Python · No API Keys · No Setup

Free ForeverMIT License Live Cloud Backend