DAEMON-AI

Detection & Active Elimination of Malicious Anomalies through Ongoing Network, Sensor & Behavioral Analysis with AI

Visit Project Website

Project Overview

The DAEMON-AI project, led by Malloc LTD in collaboration with The Cyprus Institute (CaSToRC), aims to revolutionise mobile cybersecurity by developing a behaviour-based anomaly detection system that functions in real time, entirely on device.

Traditional signature-based detection methods struggle with new, unknown threats such as sophisticated spyware, unauthorised data transmissions and encrypted traffic anomalies. DAEMON-AI addresses these limitations by using lightweight, on-device unsupervised machine learning models for anomaly detection — enabling low latency, high accuracy and robust data protection.

Integrated into the Malloc mobile protection stack, this system will elevate the application’s ability to detect unknown threats without relying on cloud infrastructure, thereby enhancing user privacy and resilience.

Objectives & Expected Results

DAEMON-AI has two general objectives:

  • Develop an innovative solution for identifying cybersecurity threats by enhancing Malloc's mobile app with a lightweight AI network that detects anomalies in encrypted network traffic and provides real-time protection — strengthening security for Malloc's users and its position in the global cybersecurity market.
  • Strengthen the long-term collaboration between Malloc and the Cyprus Institute (CYI), combining industry and academic expertise in cybersecurity, AI and high-performance computing.

Expected results:

  • An on-device behavioural anomaly detection model, built for privacy and speed — designed to flag a threat in under a second without a noticeable battery cost.
  • On-device classification of encrypted traffic types (e.g. video, audio, documents, text) at a target accuracy of at least 80%, without ever decrypting user data.
  • Integration of DAEMON-AI's detection capability into Malloc's commercial app, released to users in stages.
  • Real-world testing with at least 100 opted-in users, plus dissemination through conferences, publications and outreach events.

Key Facts

  • Acronym: DAEMON-AI – “Detection and Active Elimination of Malicious anomalies through Ongoing Network analysis with AI”
  • Funding: Funded by the Research Innovation Foundation of Cyprus under RIF/CODEVELOP/0824 for 2 years (from 01/07/2025 to 30/06/2027)
  • Coordinator: Malloc LTD
  • Partner: The Cyprus Institute (CaSToRC)
  • Focus: On-device anomaly detection, sensor & behavioural analysis, encrypted traffic monitoring, privacy-preserving AI.

Partnership

A collaborative effort between leading research institutions and industry partners, supported by Cyprus's innovation ecosystem.

The Cyprus Institute
Malloc LTD

Funding

The Project “CODEVELOP/0824/0328" is implemented under the programme of social cohesion “THALIA 2021-2027” co-funded by the European Union, through Research and Innovation Foundation (Proof of Concept for Technology / Knowhow Applications Programme).

Co-Funded by the European Union

Why It Matters

As spyware tools, surveillance techniques and encrypted threats continue to escalate in sophistication, the need for adaptive, privacy-first cybersecurity on mobile devices has never been greater. DAEMON-AI pushes the frontier by embedding unsupervised anomaly detection directly on device, ensuring your data stays local, secure and free from external inference.

For enterprises, high-risk individuals and government agencies who cannot rely on cloud-based detection alone, DAEMON-AI represents a significant leap in resilience, autonomy and trust. Malloc’s collaboration with The Cyprus Institute positions this technology at the intersection of state-of-the-art research and real-world deployment.