Inspiration

In an era of 'Passive Deception,' AI-generated media has become so pervasive that we have stopped questioning what is real. This lack of clarity creates a "Digital Trust Gap" that disproportionately affects vulnerable groups. We saw seniors being targeted by synthetic content they couldn't verify and young people struggling with the mental health impacts of AI-generated "perfection." We were inspired to build a tool that acts as a pair of 'digital glasses,' restoring transparency to the internet.

What it does

Pixel Zero is a real-time AI Feed Detector and safety tool. It monitors your browser to immediately detect AI or unsafe content and inform you.

  • Active Feed Scanning: Automatically identifies and labels AI-generated images and videos as you scroll through social media.
  • NSFW Transparency: Uses AI to detect and blur sensitive or synthetic adult content, replacing it with a "Reddit-style" reveal interaction.
  • The Lasso Tool: A precision "superpower" that allows users to draw a rectangle around any specific area of a webpage for instant, on-demand AI verification.

How we built it

  • Frontend: A Chrome Extension built with JavaScript that performs DOM injection and provides a custom UI overlay.
  • Backend: A Flask-based Python server that orchestrates complex detection tasks.
  • AI Logic: Integrated Google Gemini for advanced visual reasoning and NSFW classification, alongside custom algorithms to generate unique image hashes for real-time tracking.
  • The Pipeline: We built a custom "hand-off" chain that moves data from the browser to our API and back to the specific DOM element in milliseconds.
  • AI_or_Not API: Used the best AI detection API available to get accurate results and detect AI.

Challenges we ran into

One of the biggest hurdles was the "Data-Flow Gap"—ensuring that a detection result from the backend accurately found its way back to a specific image in a dynamically scrolling feed (like Instagram). We solved this by implementing a robust hashing system. We also faced significant challenges in UI/UX: creating an NSFW overlay that was protective but not intrusive, which led to our "Are you sure?" confirmation hover logic. Handling videos was also challenging due to inconsistencies while extracting the content from the DOM between the developer environments.

Accomplishments that we're proud of

  • Real-Time Responsiveness: Achieving near-instant feedback when a user "lassos" an image on their screen.
  • User-Centric Safety: Successfully implementing a blur-and-reveal system that empowers the user rather than just blocking content.
  • Seamless Integration: Building a tool that lives in the background without degrading the browsing experience.

What we learned

We learned that digital literacy isn't just about education; it's about tools. We discovered that by providing a simple "Trust Anchor," we can significantly reduce the cognitive load on users trying to navigate a synthetic world. Technically, we deepened our knowledge of asynchronous message passing in browser extensions and real-time image processing.

What's next for Pixel Zero

  • Custom Detection Algorithms: Want to be able to create a heatmap of the AI-generated or edited content.
  • RAG: To determine whether content is NSFW rather than just the API call.
  • Seamless Experience: Have it work across the browser altogether while keeping the UI nice and clean.
  • Fraud Detection: Make it easier to detect fraud while monitoring screens by keeping up to date about the latest scams that one can face, i.e. phishing or pretending to be someone's grandson scam.
  • Mobile Integration: Bringing the Pixel Zero safety layer to mobile browsers to protect users on the go.
Share this project:

Updates