Great post!
The community needs to be more vigilant, but it starts before even opening the article. If an account has, for example, 12 different articles on 12 different niche subjects from health, to travel, to body building, it should at the very least, raise some eyebrows.
Secondly, the personal touch goes a long way. Some of the most authentic bloggers on Steemit have great support because they inject their personality into everything they do, through photos, writing style, or sticking to their preferred niches.
Within the text, strange grammar is far more often a result of spun content than poor English in my experience. There are other methods emerging too, some of them extremely cynical, such as using special characters that look like regular letters, in order to avoid detection.
As you can see in this picture, the spun words with special characters can look identical to the real words, and only a spell checker will highlight them.
Admittedly, this method is currently quite effective. It's also giving me a lot of extra work when curating for Steemcleaners. Perhaps we will see the emergence of a robot similar to cheetah that would comment on posts with excessive special characters. Anyone want to organise a bounty for that?
Steemcleaners are catching a lot of this stuff, across all of the spinning methods, including translation, spinning, and special characters. Ideally though (IMO), plagiarised content simply could not be rewarding because the community would demand more authentic personalities.