It's simple. Remember not to copy directly from the internet. Even if you change some words, the system will detect them. You should know that the detection software links to a huge database, and all articles are included, just like Google Maps, it is a huge database. Of course, a computer is a machine after all, and it can only think physically, not like people. For example, if the copy on the computer is "Hello", we can rewrite it as "How are you recently", so that this problem can be solved simply. You copy your article into Google Translate, first translate it into English, then copy it into a dictionary, and then translate it into Chinese, so that the computer can't detect it. Finally, you can rewrite untranslatable sentences into fluent ones. It's simple. Oh, that's what I did. The first detection rate of the paper is 12%, and this 12% also includes the cited text of the paper.

Another way is: you can copy articles from books in the library, even if you copy them word by word, the computer can't detect them, because it is impossible for the computer to collect all the contents of books into the database.