A Brief Overview of Privacy-Preserving Technology Methods

Mar 22, 2018
Share this post
Sharing to FacebookSharing to LinkedInSharing to XSharing to Email

A very brief overview of data privacy software for machine learning engineers, developers, data enthusiasts, and anyone interested in learning more about data privacy.  This article will cover: symmetric encryption, asymmetric encryption, homomorphic encryption, differential privacy, and secure multi-party computation.

Symmetric Encryption

What is it? 

An alternative nomenclature for symmetric encryption is private key encryption, which means that both encryption and decryption are performed using the same key that is kept hidden from everyone but the parties that want to communicate securely. This includes block ciphers (e.g., AES-GCM) and some stream ciphers (e.g., one-time pads), among others.

What is a practical application? 

File sharing (e.g., over the HTTPS protocol) and hard drive encryption.

Pros 
  • Symmetric schemes allow for faster encryption and decryption than asymmetric ones, which will be discussed in the next section 
  • Symmetric schemes are mathematically simpler
Cons 
  • The encryption/decryption key has to be exchanged over a secure channel; i.e., a communication channel where the message being sent cannot be read or tampered with
Some seminal papers
Software
A Brief Overview of Privacy-Preserving Technologies

Asymmetric Encryption

What is it? 

Asymmetric encryption is also referred to as public key encryption, which includes the RSA cryptosystem and ElGamal, among others. These schemes require a private key to decrypt but, unlike symmetric schemes, the encryption key is public. Crucially, asymmetric encryption schemes enable two parties to securely exchange the necessary keys (e.g., for future communication secured using symmetric encryption) without worrying about the level of security of the exchange channel.

What is a practical application? 

Asymmetric and symmetric schemes are commonly combined to provide the best of both worlds. A sender may encrypt a secret key for a symmetric cipher using a receiver’s public key, then send the encrypted key to the receiver, who in turn decrypts it. Encrypted communication can then be more efficient, since both the sender and receiver now have the key to the same symmetric cipher.

Pros 
  • Removes the need for communication over a secure channel
Cons 
  • It is known to be a slow process
Some seminal papers
Software

Homomorphic Encryption

What is it? 

The concept of homomorphic encryption is said to be envisioned by Adleman, Dertouzos, and Rivest in 1978. Using homomorphic encryption schemes, a user is able to encrypt data then send the data to an untrusted server that performs computations on the encrypted data. The server then returns the result of those computations to the user. Finally, the user decrypts the result. This scheme allows one, for example, to encrypt the numbers 2 [i.e., e(2)] and 3 [i.e., e(3)] and perform e(2) + e(3), decrypt the result, and get 5. 

A Brief Overview of Privacy-Preserving Technologies and software

Certain schemes can only perform addition, others only multiplication, and others still both multiplication and addition. Notably, the Paillier cryptosystem can only perform addition and is thus said to be additively homomorphic. One can perform only multiplication using RSA, for instance, making it multiplicatively homomorphic. Lattice-based encryption schemes are a favourite, since they are additively and multiplicatively homomorphic, as well as quantum-safe.

What is a practical application? 

This data privacy software is fairly new for commercial applications. In practice, it has been used in the financial and healthcare sectors.

Pros 
  • Homomorphic encryption allows for double-blind computations. That is, a server is blind to a user’s data and the user is blind to the intricacies of the server’s algorithms
Cons 
  • Non-linear operations must rely on user-server communication, where the data to be computed on is sent to the user with some noise added to it. The user then performs the computation and sends it back to the server. 
  • Examples of such operations would be logarithms, sines, cosines, etc. Generally, a huge benefit of homomorphic encryption is its low communication costs between the user(s) wishing to perform a computation and the server performing it. 
    • Requiring user-server communication for operations which are otherwise trivial to perform limits the practicality of HE.
Some seminal papers
Software

Differential Privacy

What is it? 

Differential privacy allows one to learn general information about a population within a database or within aggregated datasets without learning anything about a specific individual. A common example of a useful task for differential privacy is using a database to determine whether smoking causes cancer. In that situation, one is discovering a general fact about a population (i.e., smokers) without requiring any individually identifiable details.

What is a practical application? 

Apple uses differential privacy to improve QuickType predictions, as well as some of their recommendations to users. Briefly, only a certain portion of a user’s true actions are sent to Apple to train their models. It is impossible for Apple to distinguish between a user’s true action and a randomized action. However, the majority of information they receive is accurate and therefore improves predictions.

Pros 

The risk of identifying a single anonymized person in a dataset can be accurately quantified as a function of the parameters used. This is mathematically provable.

Cons 

Loss of data precision. The more deidentified or noisy that a dataset is, the less useful it becomes. This is definitely not the right solution if one needs exact answers to very specific queries (e.g., how many people in a certain dataset have cancer).

Some seminal papers
Software

Secure Multi-Party Computation

What is it? 

This data privacy software is used when multiple parties have a portion of data necessary for computing the result of a function, but do not want to reveal their data to any of the other parties. Garbled circuits, or circuits which hide all information going through them except for the final output, are used for MPC.

What is a practical application? 

Deriving genomic diagnoses without revealing patient genomes.

Pros 
  • Very low computational cost
Cons 
  • Little research done on performing comparison (<, >, ==, !=) within MPC protocols.
Some seminal papers
Software
Other notable software

Join Private AI for more discussions on the best data privacy software methods on LinkedIn, Twitter, and Youtube

Acknowledgements

Thank you to Dr. Siavash Kazemian, Kelly Langlais, Michael Chiu, and Simon Emond for their feedback on this post and to Dr. Parinaz Sobhani for telling me about Apple’s great use case of differential privacy.

Data Left Behind: AI Scribes’ Promises in Healthcare

Data Left Behind: Healthcare’s Untapped Goldmine

The Future of Health Data: How New Tech is Changing the Game

Why is linguistics essential when dealing with healthcare data?

Why Health Data Strategies Fail Before They Start

Private AI to Redefine Enterprise Data Privacy and Compliance with NVIDIA

EDPB’s Pseudonymization Guideline and the Challenge of Unstructured Data

HHS’ proposed HIPAA Amendment to Strengthen Cybersecurity in Healthcare and how Private AI can Support Compliance

Japan's Health Data Anonymization Act: Enabling Large-Scale Health Research

What the International AI Safety Report 2025 has to say about Privacy Risks from General Purpose AI

Private AI 4.0: Your Data’s Potential, Protected and Unlocked

How Private AI Facilitates GDPR Compliance for AI Models: Insights from the EDPB's Latest Opinion

Navigating the New Frontier of Data Privacy: Protecting Confidential Company Information in the Age of AI

Belgium’s Data Protection Authority on the Interplay of the EU AI Act and the GDPR

Enhancing Compliance with US Privacy Regulations for the Insurance Industry Using Private AI

Navigating Compliance with Quebec’s Act Respecting Health and Social Services Information Through Private AI’s De-identification Technology

Unlocking New Levels of Accuracy in Privacy-Preserving AI with Co-Reference Resolution

Strengthened Data Protection Enforcement on the Horizon in Japan

How Private AI Can Help to Comply with Thailand's PDPA

How Private AI Can Help Financial Institutions Comply with OSFI Guidelines

The American Privacy Rights Act – The Next Generation of Privacy Laws

How Private AI Can Help with Compliance under China’s Personal Information Protection Law (PIPL)

PII Redaction for Reviews Data: Ensuring Privacy Compliance when Using Review APIs

Independent Review Certifies Private AI’s PII Identification Model as Secure and Reliable

To Use or Not to Use AI: A Delicate Balance Between Productivity and Privacy

To Use or Not to Use AI: A Delicate Balance Between Productivity and Privacy

News from NIST: Dioptra, AI Risk Management Framework (AI RMF) Generative AI Profile, and How PII Identification and Redaction can Support Suggested Best Practices

Handling Personal Information by Financial Institutions in Japan – The Strict Requirements of the FSA Guidelines

日本における金融機関の個人情報の取り扱い - 金融庁ガイドラインの要件

Leveraging Private AI to Meet the EDPB’s AI Audit Checklist for GDPR-Compliant AI Systems

Who is Responsible for Protecting PII?

How Private AI can help the Public Sector to Comply with the Strengthening Cyber Security and Building Trust in the Public Sector Act, 2024

A Comparison of the Approaches to Generative AI in Japan and China

Updated OECD AI Principles to keep up with novel and increased risks from general purpose and generative AI

Is Consent Required for Processing Personal Data via LLMs?

The evolving landscape of data privacy legislation in healthcare in Germany

The CIO’s and CISO’s Guide for Proactive Reporting and DLP with Private AI and Elastic

The Evolving Landscape of Health Data Protection Laws in the United States

Comparing Privacy and Safety Concerns Around Llama 2, GPT4, and Gemini

How to Safely Redact PII from Segment Events using Destination Insert Functions and Private AI API

WHO’s AI Ethics and Governance Guidance for Large Multi-Modal Models operating in the Health Sector – Data Protection Considerations

How to Protect Confidential Corporate Information in the ChatGPT Era

Unlocking the Power of Retrieval Augmented Generation with Added Privacy: A Comprehensive Guide

Leveraging ChatGPT and other AI Tools for Legal Services

Leveraging ChatGPT and other AI tools for HR

Leveraging ChatGPT in the Banking Industry

Law 25 and Data Transfers Outside of Quebec

The Colorado and Connecticut Data Privacy Acts

Unlocking Compliance with the Japanese Data Privacy Act (APPI) using Private AI

Tokenization and Its Benefits for Data Protection

Private AI Launches Cloud API to Streamline Data Privacy

Processing of Special Categories of Data in Germany

End-to-end Privacy Management

Privacy Breach Reporting Requirements under Law25

Migrating Your Privacy Workflows from Amazon Comprehend to Private AI

A Comparison of the Approaches to Generative AI in the US and EU

Benefits of AI in Healthcare and Data Sources (Part 1)

Privacy Attacks against Data and AI Models (Part 3)

Risks of Noncompliance and Challenges around Privacy-Preserving Techniques (Part 2)

Enhancing Data Lake Security: A Guide to PII Scanning in S3 buckets

The Costs of a Data Breach in the Healthcare Sector and its Privacy Compliance Implications

Navigating GDPR Compliance in the Life Cycle of LLM-Based Solutions

What’s New in Version 3.8

How to Protect Your Business from Data Leaks: Lessons from Toyota and the Department of Home Affairs

New York's Acceptable Use of AI Policy: A Focus on Privacy Obligations

Safeguarding Personal Data in Sentiment Analysis: A Guide to PII Anonymization

Changes to South Korea’s Personal Information Protection Act to Take Effect on March 15, 2024

Australia’s Plan to Regulate High-Risk AI

How Private AI can help comply with the EU AI Act

Comment la Loi 25 Impacte l'Utilisation de ChatGPT et de l'IA en Général

Endgültiger Entwurf des Gesetzes über Künstliche Intelligenz – Datenschutzpflichten der KI-Modelle mit Allgemeinem Verwendungszweck

How Law25 Impacts the Use of ChatGPT and AI in General

Is Salesforce Law25 Compliant?

Creating De-Identified Embeddings

Exciting Updates in 3.7

EU AI Act Final Draft – Obligations of General-Purpose AI Systems relating to Data Privacy

FTC Privacy Enforcement Actions Against AI Companies

The CCPA, CPRA, and California's Evolving Data Protection Landscape

HIPAA Compliance – Expert Determination Aided by Private AI

Private AI Software As a Service Agreement

EU's Review of Canada's Data Protection Adequacy: Implications for Ongoing Privacy Reform

Acceptable Use Policy

ISO/IEC 42001: A New Standard for Ethical and Responsible AI Management

Reviewing OpenAI's 31st Jan 2024 Privacy and Business Terms Updates

Comparing OpenAI vs. Azure OpenAI Services

Quebec’s Draft Regulation Respecting the Anonymization of Personal Information

Version 3.6 Release: Enhanced Streaming, Auto Model Selection, and More in Our Data Privacy Platform

Brazil's LGPD: Anonymization, Pseudonymization, and Access Requests

LGPD do Brasil: Anonimização, Pseudonimização e Solicitações de Acesso à Informação

Canada’s Principles for Responsible, Trustworthy and Privacy-Protective Generative AI Technologies and How to Comply Using Private AI

Private AI Named One of The Most Innovative RegTech Companies by RegTech100

Data Integrity, Data Security, and the New NIST Cybersecurity Framework

Safeguarding Privacy with Commercial LLMs

Cybersecurity in the Public Sector: Protecting Vital Services

Privacy Impact Assessment (PIA) Requirements under Law25

Elevate Your Experience with Version 3.5

Fine-Tuning LLMs with a Focus on Privacy

GDPR in Germany: Challenges of German Data Privacy (Part 2)

Comply with US Executive Order on Safe, Secure, and Trustworthy Artificial Intelligence using Private AI

How to Comply with EU AI Act using PrivateGPT