Avast Research Lab

The Avast Research Lab runs innovative projects in all areas of digital safety, from advanced threat detection, to delivering better privacy and identity protection, and much more. We employ state-of-the-art AI solutions to counter the ever accelerating growth of emergent threats through a combination of in-house expertise, academic cooperation, and publicly available research.

Consumer security in the AI-driven world

The goal of our Research program is to explore and advance cutting-edge research in the fields of AI and machine learning for cybersecurity and online privacy protection. We develop solutions that are impactful to all digital citizens in today’s connected world.

We believe that wherever you live, wherever you go, and whatever you do, you have the right to live freely and safely online.

“The Avast AI Research Lab is helping Avast and the whole consumer cybersecurity community to understand future challenges that internet users will be facing in the digital age. This includes emerging threats in a fast changing landscape, such as AI-powered cyber attacks, but also in the space of users’ privacy and ownership of private data. Consumer identity management, manipulation prevention, and protection are all other big challenges. I’m excited that we were able to build such a diverse and forward-looking organization within Avast. And I’m looking forward to working with the team.”
Michal Pechoucek
CTO at Avast and AI Professor at Czech Technical University in Prague

Research areas

Robust & explainable AI

We make AI models more reliable, easier to understand, and easier to use by humans — especially with regard to machine data. We build in-house and reusable models for efficient learning with structured machine JSON type data.

Autonomous defense

We strive to change cybersecurity at its core. We build AI models to automate the bulk of analytics work, and we build systems that scale with the accelerating growth of the threat landscape.

Secure internet

We understand the internet as the digital fabric that surrounds us; the digital environment we live in. The future internet iteration made up of persistent, shared, 3D virtual spaces called the metaverse will require the security stack to be completely rebuilt.

Personalized security and privacy

We tackle personalized security and privacy as the next frontier in consumer security. General security measures best mitigate mass deployed threat vectors. Personalized security and privacy aims to protect against threats targeting individuals and devices.

Massively distributed computing

We protect hundreds of millions of users across the globe. We build technology and systems capable of serving all of our users at lightning speed.

Digital freedom frontiers

We believe users should keep total control of their digital lives. We develop privacy-enhancing and digital identity management technologies. We invest in attention-stealth detection and misinformation detection research.

Collaboration

Publications

Transparent Microsegmentation in Smart Home IoT Networks

Amr Osman, Stefan Köpsell, Thorsten Strufe (TU Dresden) and Armin Wasicek (Avast)
Workshop on Hot Topics in Edge Computing collocated with 2020 USENIX Annual Technical Conference, June 2020, Paper

Discovering Imperfectly Observable Adversarial Actions using Anomaly Detection

Olga Petrova (Avast), Karel Durkota (Czech Technical University in Prague), Galina Alperovich (Avast), Karel Horak (Czech Technical University in Prague), Michal Najman, Branislav Bosansky, Viliam Lisy (Avast) AAMAS, May 2020, Extended abstract

Geost botnet. The discovery story of a new Android banking trojan from an OpSec error

‍Sebastián García (Czech Technical University in Prague), Maria Jose Erquiaga (UNCUYO University), Anna Shirokova (Avast) Virus Bulletin International Conference, October 2019, Paper

Featured blogs

Podcast

What does the future hold for cybersecurity

by AVAST AI RESEARCH LAB We’re speaking about AI and cybersecurity now and what it...

November 3, 2021

Podcast

Federated learning with Fabrizio Biondi and Richard Chow

by AVAST AI RESEARCH LAB Is decentralized machine learning the ultimate way forward in the...

October 21, 2021

Article

Is it malware or clean? Well, it depends on a plethora of diverse features

by AVAST AI RESEARCH LAB Avast researchers use a general feature-blind learning framework for fast...

October 18, 2021