A GIF is a specific image format that uses a specific set of techniques to compress an image and represent multiple frames. Those techniques are very cheap to decompress and thus was very good for the early internet when computers weren’t very fast.
Modern video formats use several different techniques together for better image compression, most (all?) of which the GIF image format does not support. They also require significantly more computation to decompress, with several video formats actually using specialized hardware on your device or computer to play back as they could otherwise be too expensive to play back in real time!
The basic explanation of the GIF compression format is that it uses indexed color. A table of up to 256 colors is stored for each GIF with the pixels of the image only storing what color index to use. If multiple pixels in a row use the same color the format just stores the one color index and how many pixels after that use the same color. You can think of it as a digital paint by numbers.
For multiple frames of animation, each frame generally only stores information about pixels that changed from the previous frame, and uses a single “transparent” color for the other pixels. This can often mean much of each frame after the first is almost blank if not much changed.
Then there’s a little information about how long to display each frame and how many times to loop the animation, and that’s kind of it.
It is possible to change or add to the GIF format, looping wasn’t originally a feature for example, but there’s little reason for anyone to spend the time doing so when there are other formats that already exist. It would also require getting browser and content authoring tool developers to support those changes.