The latest Search Off the Record podcast gives a helpful explanation of lazy loading (lives up to its title). Here's the gist:

• Lazy loading defers loading non-critical resources (mostly images & videos) on a webpage until they are needed (e.g. when a user scrolls them into view).

◦ Without lazy loading, all resources are loaded as soon as the page is opened.

• *Benefits*:

◦ Faster page load times & CWV because initial load is smaller

◦ Less battery power needed on mobile devices (!)

◦ Can save money if a resource relies on API calls (e.g. stock ticker) cuz they won't load if they're not in view

• *Native HTML implementation is best*: slap a `loading="lazy"` attribute on images & iframes (modern browsers support this). WP does by default.
• Custom JS libraries were used before native HTML was possible & sometimes still are - but *if not implemented right, they can hurt indexing* (e.g. if the library doesn't swap a custom attribute like `data-src` for the standard `src`, Googlebot won't see the image URL)
• *You can check if lazy loading is done right in GSC*: URL inspection tool > inspect URL > view rendered HTML > make sure all lazy loaded content (probably images) is in the rendered HTML using the `src` attribute in the image tag
• *Infinite scroll looks similar but it's not the same thing*: lazy loading defers loading of part of the page content, but infinite scroll loads additional content onto the page
• *Decorative images are better off in the CSS* so they're not part of indexing