Civilsphere AI VPN: v0.1-alpha pre-release

AI-VPN-Logo-Idea-4_bl-up_bg-white_1500x1500.jpg

Civil society organizations, activists, journalists and human rights defenders rely on digital devices to do their jobs, and are often in the field without the protection of firewalls, IDS, or any other advanced threat protection more commonly available to corporate users. How can these individuals identify if their devices are at risk while working in the field?

Since 2018, our Emergency VPN service has been leveraging VPN technologies to provide users with a security assessment of their devices’ network traffic tailored to identify digital threats. Our new AI VPN aims to take this work even further.

The AI VPN is a free software tool that provides users with an automatic network security assessment of their network traffic. Users can request a new VPN profile via email, connect to it for a predefined amount of time (hours to days), and receive a full report after their VPN profile expires. 

The AI VPN aims to enable organizations to deploy their own instance without the need to rely on third-parties with their data.

AI VPN v0.1-alpha pre-release

This project is still in its early development, however we are happy to have an alpha pre-release already available for testing. This first release includes some of the AI VPN core functionality: 

  • Automatic generation of VPN accounts

  • Automatic network traffic captures per account

  • Automatic revocation of VPN accounts

  • Automatic report generation of the captured traffic per account

  • Email is currently the supported method for sending VPN requests and obtaining the network traffic analysis reports.

  • VPN supported technologies currently include OpenVPN

  • AI VPN currently handles up to 253 simultaneous users.

This version was tested using Docker v20.10.6 and docker-compose v1.28.5.

The AI VPN documentation can be accessed at https://aivpn.readthedocs.io. The application can be downloaded in our GitHub repository https://github.com/stratosphereips/AIVPN.

Acknowledgements

This project was funded through the NGI0 PET Fund, a fund established by NLnet with financial support from the European Commission's Next Generation Internet programme, under the aegis of DG Communications Networks, Content and Technology under grant agreement No 825310.

This project was developed by the Stratosphere Laboratory, Artificial Intelligence Centre, Faculty of Electrical Engineering, Czech Technical University in Prague (2020-2021).