JevMade Sign in
← Back to guides

JevMade field notes / Architecture and workflow guide

How to build an AI tool that finds and removes ads

This guide explores a test project that uses an AI tool to spot and remove website ads. It shows how standard software handles the actual removal while the AI only decides if something is an advertisement.

Original by realZachiClassification

Listen to this guide

JevMade’s plain-English explanation

0:00 /

Our summary

This project is a custom web browser tool that removes advertisements from web pages. A person might use this to see how an AI tool can judge parts of a website as you browse, rather than using it as a complete replacement for standard ad blockers.

Standard software first searches the page for likely ads based on labels like the word sponsored. It sends these options to Jev, an AI tool that chooses from options rather than writing an answer. If Jev is confident an item is an ad, the software removes it.

This setup requires a private access code called an API key, which costs money for every page scanned. It will miss some ads and cannot block hidden tracking software. It is mostly useful for people who want to test how AI handles simple choices safely.

Key takeaways

  1. Use standard software to find likely ads before paying an AI assistant to check them.
  2. Keep the actual removal rules in your software so the AI cannot delete random items.
  3. Test the tool with a highlight mode to catch mistakes before letting it delete anything.

The author reported the testing results and removal speeds.

Repository README and documentation

Read the original guide Opens the author’s site in a new tab.

Keep this for later

Sign in to bookmark experiments, guides and videos, and keep notes only you can see.

Continue to sign in

We’ll bring you back to this listing.