Inspiration
With the rapid rise of AI-generated content, deepfakes and online scams are becoming increasingly convincing and widespread. People often struggle to distinguish between real and manipulated content, leading to financial loss and misinformation. We were inspired to build a solution that acts at the moment of interaction, helping users verify content before they trust it.
What it does
EmpowerNet is an AI-powered deepfake and scam detection platform that works in real time. It analyzes images, videos, and text using multiple AI models and provides instant risk scores with explanations. Through a browser-based approach, it protects users directly on the platforms they already use, without requiring technical knowledge.
How we built it
We built EmpowerNet by integrating multiple AI models for different detection tasks—computer vision models for deepfake detection and NLP models for scam and phishing analysis. The backend handles model orchestration and scoring, while the frontend delivers a simple and intuitive user experience. We also experimented with onchain verification (testnets) to create tamper-proof records of detection results.
Challenges we ran into
One of the biggest challenges was handling multiple heavy AI models efficiently, especially under limited resources. We had to optimize memory usage and implement smarter loading strategies. Another challenge was ensuring real-time performance while maintaining accuracy, as well as designing outputs that are easy for non-technical users to understand.
Accomplishments that we're proud of
Built a multi-modal detection system (image, video, text) Achieved real-time analysis with explainable results Developed a browser-integrated concept for seamless protection Successfully tested onchain verification concepts on testnets Created a working prototype within a short timeframe
What we learned
We learned how to combine multiple AI systems into a unified product, optimize performance under constraints, and design for real-world usability rather than just technical accuracy. We also gained insights into user trust, explainability, and the importance of timing in security products.
What's next for EmpowerNet – AI Deepfake and Scam Detection Platform
Our next steps are to improve model accuracy, expand detection capabilities, and fully launch the browser extension for real-world usage. We also plan to scale our onchain verification layer, explore partnerships with platforms, and move toward active user growth and monetization.
Log in or sign up for Devpost to join the conversation.