Blog icon

Cybersecurity for a secure Australia

A safe and secure nation is underpinned by strong cyber research capabilities. 

The complexity, density and distribution of digital products and networks is rapidly growing. Virtually every Australian owns a laptop, smart phone, voice assistant or smart watch, and uses a computer or digital device at work. As this adoption continues, so too does the spread, frequency and severity of cyber attacks, making cybersecurity a critical consideration.

With cybercrime a growing reality for Australians, we’re working to develop resilient and trustworthy systems that ensure private data is kept secure and safe. We’re using cutting-edge technology and research to build innovative and adaptable cybersecurity products and services that will keep Australia a step ahead. 

Featured cybersecurity research

Digitally disappearing

A cyber attack that allows objects to digitally disappear on video despite being clearly visible could set in motion the next generation of undetectable crime. As machine learning detection systems are trained to analyse video footage and identify people, an ‘invisible cloak’ attack can be used against these systems to identify what can be worn to remain undetected. Here's how our researchers helped discover the attack and what can be done to prevent it.

Cyber for society

Magda data catalog

With Magda, data analysts, scientists and engineers can easily find useful data with powerful discovery features and make data-informed decisions with confidence.

Enhancing Speech Emotion Recognition using Machine Learning Techniques

Recognising human emotion in technology has always been fascinating work for data scientists. CSIRO’s Data61 is advancing the science of Speech Emotion Recognition (SER).

Metrics and frameworks for privacy risk assessments

How to confidently measure and understand privacy and confidentiality risks in data sharing.

Creating safe and secure systems

This common action is putting you at risk of being deepfaked

What is a deepfake? Deepfaking is the act of using artificial intelligence and machine learning technology to produce or alter video, image or audio content using the sequence of the original to create a version of something that didn’t occur.

How you can a-Void a voice spoofing attack

Siri, Alexa, Bixby and Google Home all allow people to use voice commands to quickly shop online, make phone calls, send messages, control smart home appliances, and access banking services. Because of these personal and security–critical commands, voice assistants are lucrative targets for attackers to exploit.

Why we are vaccinating algorithms against adversarial attacks

As the use of artificial intelligence (AI) and machine learning across our society and economy is increasing, so too is the threat of these technologies being manipulated by bad actors. Our researchers have developed a world-first set of techniques, similar to the vaccination process, to inoculate machine learning models against such attacks.

Protecting and preserving privacy

PhishZip: A new compression-based algorithm for detecting phishing websites

Phishing has grown significantly since 2010 and is currently experiencing a huge surge during the COVID-19 pandemic. The dynamics of phishing attacks makes it highly challenging to implement a robust phishing detection system. However, a novel method is changing that.

What is data privacy and why it's so important

Over the last few years, data privacy and its lifecycle has become an increasing concern of consumers and businesses alike, with both adopting and trying to keep up with the rapidly changing digital environment

Web of worries

In a mere 20 years, the Internet has evolved from a fledgling ‘network of networks’ developed by Tim Berners-Lee to the all-encompassing behemoth that it is today. But the World Wide Web is a tangled mass of interconnected services, and our researchers have spotlighted a fundamental flaw in its design that has significant implications on privacy and security.

Understanding and navigating risk

Decade of cybersecurity report could help reduce the future of cyber crime

Known as ‘FinalBlacklist’, the paper collates 51.6 million mal-activity reports dating back to 2007, aggregating 662,000 unique IP addresses worldwide, which were categorised using machine learning techniques into six classes of mal-activity: Malware, Phishing, Fraudulent Services, Potentially Unwanted Programs, Exploits and Spamming.

Eight tips on staying cyber safe during COVID-19

As the majority of us continue to work from home and we use our phone or computer to stay in touch with family, friends and colleagues, ensuring your cyber safety is not only crucial but possible with these eight tips.

Making anonymous data truly anonymous

Even though data is submitted anonymously, it only takes three bits of crucial information to re-identify someone. To counter this, we've developed a system to measure the risk of re-identification and prevent it from occuring.

More cyber research

Contact us

Find out how we can help you and your business. Get in touch using the form below and our experts will get in contact soon!

CSIRO will handle your personal information in accordance with the Privacy Act 1988 (Cth) and our Privacy Policy.


This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.

First name must be filled in

Surname must be filled in

I am representing *

Please choose an option

Please provide a subject for the enquriy

0 / 100

We'll need to know what you want to contact us about so we can give you an answer

0 / 1900

You shouldn't be able to see this field. Please try again and leave the field blank.