Guessing Amazon image URLs using GitHub Copilot

I was experimenting with the new Readwise export API and it gave me back the following JSON:

{ "user_book_id": 15433610, "title": "Screenwriting: The Sequence Approach", "author": "Paul Joseph Gulino", "readable_title": "Screenwriting", "source": "kindle", "cover_image_url": "https://images-na.ssl-images-amazon.com/images/I/51UMeAAJNRL._SL75_.jpg", "unique_url": null, "book_tags": [], "category": "books", "readwise_url": "https://readwise.io/bookreview/15433610", "source_url": null, "asin": "B00F9476Y0", "highlights": [] }

The image URL https://images-na.ssl-images-amazon.com/images/I/51UMeAAJNRL._SL75_.jpg produced a tiny little image - this one:

A tiny picture of a book cover

I wanted to get back a bigger version of that image. On a hunch, I popped open a VS Code window running GitHub Copilot and typed this:

https://images-na.ssl-images-amazon.com/images/I/51UMeAAJNRL._SL75_.jpg # That image but bigger:

It autocompleted for me:

https://images-na.ssl-images-amazon.com/images/I/51UMeAAJNRL.jpg

Which is indeed a larger version of the image!

Screenwriting: The Sequence Approach - a much bigger book cover image, with legible text

I typed # and hit autocomplete again, to see what would happen, and got this:

# That image but smaller: https://images-na.ssl-images-amazon.com/images/I/51UMeAAJNRL._SL50_.jpg

An even tinier picture of a book cover

Another correct guess. GPT-3/Copilot clearly includes training data that has seen these URLs before.

I typed # one more time and it autocompleted again to:

# That image but with a different format: https://images-na.ssl-images-amazon.com/images/I/51UMeAAJNRL._SL50_.gif

... and this time that URL is a 404. Copilot made this one up!

Here's the full flow in VS Code:

Animation showing the above prompts being executed in VS Code

Created 2022-10-15T15:10:24-07:00, updated 2022-10-15T15:27:45-07:00 · History · Edit