Convert a PDF to black and white

That is a setting on one machine, and the next person to open the file never sees it; send the file to a print shop and they have no idea what you chose. This writes the grey into the file itself, so it is grey on any screen and at any printer.

No. Nothing is re-rendered at any point. Colours in a PDF are written as instructions inside the page, and all this does is rewrite those instructions. Text stays selectable, copyable and searchable, drawings stay vector and sharp at any zoom, and the fonts are the fonts the document came with.

When the file has a pale saturated colour in it, such as a yellow highlight or a sky-blue heading. A natural conversion measures how bright the colour is, and yellow is bright, so it comes out an almost invisible light grey on paper. Darker pulls down saturated colour only: white, black and any grey already in the document have no saturation and are left exactly as they were, so the page background stays white.

They are converted too, as long as Pictures too is on, which it is by default. A grey picture stores one channel instead of three or four, so the file usually comes out smaller as well. Turn the option off and only the text and drawings lose their colour while photographs keep theirs.

Three things, all rare in ordinary documents, and all of which come out looking like the original rather than broken: gradients and shadings, which carry their colour in a function rather than in an instruction; separation and spot inks, which are already a single ink and where a guess would be worse than leaving them; and inline images, which are copied through byte for byte because parsing them is the riskier choice.

No. All of the processing happens inside your browser on your own device, and the file never leaves it. You can disconnect from the internet after the page loads and the tool still works.