The Internet Archive is an useful repository for non-controversial public domain materials like old public domain books and other media. To a lesser extent, it is useful for non-controversial web material like personal hobby websites or old local news articles.
For anything beyond that, that is now at least the second documented instance after Taylor Lorenz that Archive.org cannot be trusted not to cave in. I wouldn't be surprised if there were more examples that simply went under the radar so far.
Archive.org also respects robots.txt - web material can retro-actively vanish when the person controlling the content inserts a robots.txt file.
Archive.is/.ph/.md hasn't removed anything so far, but no one knows who is behind them and pouring so many resources into mirroring web pages, and for what purpose. If it is really important, save it locally - you never know when it will be more convenient for the people behind Archive.is to memory hole something, or if they will simply vanish over night without any explanation the same way they have appeared.