Great tool because...

Some educators find this tool very useful in the EFL classroom because it is a good opportunity for students to do a writing task in an up-to-date way. For example, an activity could be that the teacher provides the first message of a conversation, and students have to come up with a proper response. Another possibility is to get learners to think of two famous people and then create a dialogue between them. These are just some of the possibilities provided by Peachey (2018).

However...

It is argumented that there are certain risks when using these sort of Apps, because, as the messages are fake, information can be misleading. According to Hu et al. (2024) Apps of this kind have been used in different areas to help cybercriminals "evidence" fake information. Coming back to the EFL sphere, other issues that these apps may entail are: Faking messages to humiliate or get others in trouble or fabricating messages from a friend to cover up inappropriate plans.

Here's a brief tutorial on how to use it:

Pros

As Chaka (2023) explains, "With the rising number of generative AI-powered LLM chatbots, there are growing concerns about the risks these chatbots pose to academic integrity by academics and educational institutions" (p.3). This is why, to address this concern, online AI detection tools are very helpful.

For educators, this tool is great because it allows them to check whether a text has been AI-generated or not, in only a few minutes. Besides, by being sure that a text was written by students themselves, the teacher can correct students' work according to their true abilities.

One specific characteristic of Copyleaks is that compared to other AI detection tools, this one is highly accurate. Gewirtz (2025) carried out a study in which ten AI content detectors were tested, and copyleak was in the third position according to accuracy, which means that this tool is among the best ones nowadays.

Cons

However, it should be considered that only checking a text trough an AI tool is not proof enough to detect whether a text was written with AI. Some users on the platform “Reddit” –both students and teachers- have claimed that they had tested a text written by them with these kind of tools, and they got 100% AI. This is why if we use these tools, we should be aware of their margin of error and try to develop strategies to make sure we are not falsely assuming a text has been written with AI when it is not the case.

Studies on the topic support this view as well. As Madhav et al. (2024) state, "[...] a plagiarism checker may produce false positives or false negatives i.e., mark the non-AI generated code as human-written and vice-versa, which is not valid" (p.63). 

How can you use it? You can check out this tutorial