Matt Wolfe

Trying To Solve The Biggest AI Problem

Sep 9, 2026 20 min
artificial intelligencecodingsoftware development
Watch on YouTube Follow Matt Wolfe on Rundown — free

Summary

AI summaries can be incomplete or wrong. Verify anything important against the original video.

The creator attempts to build an AI-based video verification tool to identify synthetic content. He iterates through several development stages, testing models like Gemini and SightEngine to filter AI-generated videos versus real footage.

In this video, the creator addresses the proliferation of AI-generated 'slop' online and builds a detection application to help identify such content. He begins by prompting ChatGPT to map out an architecture, initially using Google’s Gemini and SynthID models for detection. Finding those methods inaccurate, he refines the tool through extensive testing, coding, and eventually integrating the professional SightEngine API. The creator demonstrates the final tool's ability to scan Instagram and X video URLs and output reports containing visual findings and confidence indicators. Despite successful integration, he concludes that even advanced AI models still struggle with definitive detection, highlighting the limitations of current AGI claims.

Steps to follow

Locked
  1. 1
  2. 2
  3. 3
  4. 4
  5. 5

Key Points

Locked

Worth watching if: You are interested in building AI-integrated software tools, understanding how to use LLMs to generate project architectures, or learning about the current limitations of AI-content detection.

Sign in to unlock the full extract

Every claim, key point, and timestamp for this Matt Wolfe video — plus a daily email of every channel you follow.

Sign in with Google

No credit card. Free tier forever.

Watch on YouTube