Google has announced that its personal agent Gemini Spark can from now on tidy up your Google Photos library. It can process images, assemble albums, create shared albums of the best photos on its own, turn a photographed concert flyer into a calendar entry and carry out multi-step tasks.

The feature was announced by Shimrit Ben-Yair, head of Google Photos, in a post on X on Thursday evening, with a sentence that unintentionally says everything: "I have always dreamed of an agent that would help me make the most of my 143,206 photos and videos." So the product solves a problem the same platform created earlier - free and unlimited uploading that filled phones with photos nobody will ever open again.

The feature rolls out over the coming weeks, but only for Gemini AI Pro and Ultra subscribers in the US and only in English. When or whether it reaches other markets, the company did not say. The reader here, as usual, reads the announcement today and gets the feature whenever somebody in Mountain View remembers.

The strangest thing about the whole announcement is that the industry itself is starting to admit it has not sold the story. OpenAI chief executive Sam Altman told Bloomberg this week that the industry has "done a terrible job" of explaining the benefits of the technology, and that this is exactly why there is resistance among people around the world. It is hard not to notice that announcements of this kind are part of the same problem - they make assembling an album easier, and it is not as if that was hard.

This is a competitive reflex, not a strategy. Every AI upgrade has to be announced immediately, however small, because the competitor is announcing theirs. The result is an endless stream of announcements about features no user asked for, instead of one story about what actually changed. To use it you first have to connect Google Photos to Gemini, switch on Spark in the top corner of the app and write a request. Three steps for something that used to be called - make an album.