SecureAIML 🔐

Enterprise-Ready AI Model Security built on OpenSSF Model Signing

SecureAIML is the "Stripe for model security" - making enterprise-grade AI model protection accessible, user-friendly, and production-ready for every organization.

Why SecureAIML?

In the era of AI/ML, model security is critical. SecureAIML wraps the powerful OpenSSF Model Signing standard with an intuitive, enterprise-ready interface that makes securing your ML models as simple as:

Installation

pip install secureaiml

Quick Start

from secureml import SecureModel
import joblib

# Load your model
model = joblib.load("model.pkl")

# Secure it in one line
secure_model = SecureModel(model)
secure_model.sign_and_save("model.sml", identity="[email protected]")

# Load and verify
verified_model = SecureModel.load("model.sml", verify=True)
predictions = verified_model.predict(X_test)

Features

🎯 Universal Model Support

• Traditional ML: XGBoost, scikit-learn, LightGBM, CatBoost
• Deep Learning: PyTorch, TensorFlow, JAX, Keras
• Large Language Models: HuggingFace Transformers, GGUF, SafeTensors
• Computer Vision: ONNX, CoreML, TensorRT, OpenVINO
• Audio/Speech: Whisper, Wav2Vec, SpeechT5
• Multimodal: CLIP, DALL-E, GPT-4V, BLIP

🔒 Built on OpenSSF Standards

• Full integration with OpenSSF Model Signing
• Leverages Sigstore infrastructure
• Industry-standard cryptographic signing
• Keyless signing with OIDC
• Transparent and verifiable signatures

🏢 Enterprise Features

• Hardware Security Module (HSM) integration
• Cloud KMS support (AWS KMS, Azure Key Vault, GCP Cloud KMS)
• Advanced fingerprinting with Merkle trees
• Multi-signature workflows
• Compliance frameworks: SOC2, ISO27001, FIPS 140-2, HIPAA, GDPR
• Comprehensive audit trails and forensics

⚡ Developer-Friendly

• Simple, intuitive Pythonic API
• Auto-detection of model types
• Minimal configuration required
• Works with existing ML workflows
• Extensive documentation and examples

Basic Usage

from secureml import SecureModel
import joblib

# Train your model (any framework)
from xgboost import XGBClassifier
model = XGBClassifier()
model.fit(X_train, y_train)

# Secure it
secure_model = SecureModel(model)
secure_model.sign_and_save(
    "fraud_detection_model.sml",
    identity="[email protected]",
    version="2.0.0",
    description="Production fraud detection model"
)

# Load and verify
model = SecureModel.load("fraud_detection_model.sml", verify=True)
if model.is_verified:
    predictions = model.predict(X_test)

Enterprise Usage

from secureml.api.advanced import AdvancedSecureModel
from secureml.utils.config import SecurityConfig, SecurityLevel, ComplianceFramework

# Configure enterprise security
config = SecurityConfig.from_level(SecurityLevel.ENTERPRISE)
config.enable_fingerprinting = True
config.enable_merkle_trees = True
config.compliance_frameworks = [ComplianceFramework.SOC2, ComplianceFramework.ISO27001]

# Create advanced secure model
advanced = AdvancedSecureModel(model, config=config)

# Sign with AWS KMS
advanced.add_signature(
    identity="[email protected]",
    use_cloud_kms=True,
    kms_key_id="arn:aws:kms:us-east-1:123456789:key/abc-def",
    cloud_provider="aws"
)

# Validate compliance
compliance_report = advanced.validate_compliance(
    frameworks=[ComplianceFramework.SOC2, ComplianceFramework.HIPAA],
    generate_report=True,
    report_path="compliance_report.json"
)

print(f"Compliance Status: {compliance_report['overall_status']}")

Architecture

SecureML is built as an enhancement layer on top of OpenSSF Model Signing:

┌─────────────────────────────────────────────────────┐
│            Your Application                         │
└─────────────────────────────────────────────────────┘
                      ↓
┌─────────────────────────────────────────────────────┐
│            SecureML API Layer                       │
│  • Simple API  • Advanced API  • CLI                │
└─────────────────────────────────────────────────────┘
                      ↓
┌─────────────────────────────────────────────────────┐
│         SecureML Enterprise Features                │
│  • HSM/KMS  • Compliance  • Audit  • Forensics      │
└─────────────────────────────────────────────────────┘
                      ↓
┌─────────────────────────────────────────────────────┐
│         OpenSSF Model Signing (Core)                │
│         Sigstore Infrastructure                     │
└─────────────────────────────────────────────────────┘

Security Levels

SecureML provides four security levels to match your needs:

Level	Use Case	Features
BASIC	Development, testing	OpenSSF signing only
STANDARD	Production deployments	+ Fingerprinting, audit logging
ENTERPRISE	Regulated industries	+ Merkle trees, threat detection, compliance
MAXIMUM	High-security environments	+ Encryption, forensics, multi-sig

Compliance Support

SecureML helps you meet regulatory requirements:

• SOC 2: System and Organization Controls
• ISO 27001: Information Security Management
• FIPS 140-2: Cryptographic Module Validation
• HIPAA: Healthcare data protection
• GDPR: EU data protection

Documentation

• 📦 PyPI Package - Official package on PyPI
• 🚀 Quick Start Guide - Get started in 5 minutes
• 📚 Installation Guide - Installation instructions
• 📖 Usage Guide - Comprehensive usage documentation
• 🔒 Watermarking Features - Model watermarking guide
• 🛡️ Threat Model - Security analysis and limitations
• 🔗 OpenSSF Integration - OpenSSF Model Signing integration

Examples

XGBoost Model

from secureml import SecureModel
import xgboost as xgb

model = xgb.XGBClassifier()
model.fit(X_train, y_train)

secure_model = SecureModel(model)
secure_model.sign_and_save("xgb_model.sml", identity="[email protected]")

PyTorch Model

import torch
from secureml import SecureModel

model = torch.nn.Sequential(...)
torch.save(model.state_dict(), "model.pth")

secure_model = SecureModel.load_from_path("model.pth")
secure_model.sign_and_save("pytorch_model.sml", identity="[email protected]")

HuggingFace Model

from transformers import AutoModel
from secureml import SecureModel

model = AutoModel.from_pretrained("bert-base-uncased")
model.save_pretrained("./my_model")

secure_model = SecureModel.load_from_path("./my_model")
secure_model.sign_and_save("bert_model.sml", identity="[email protected]")

Installation Options

# Basic installation
pip install secureaiml