By and large, no.
The thing to remember is: they aren't being scraped "off the internet" but rather from somebody's site -- and if that site has a ToS which says the material is copyright, and/or has any other restrictions -- or there's a robots.txt that plainly forbids all form of scraping -- well, you know exactly what to do from there. Right?