Full disclosure: This article was not written by an AI ?

It should be about time for us as a society to agree on some ethics regarding the use of AI-written texts.

When is it acceptable and when is it not? This fundamental question seems to be largely ignored these days.

AIs can offer great help during the writing process, but we should bear in mind that their overuse can do more harm than good. That’s because of the fact that in communications, whether verbal or written, you often reveal much more than you intend to. There’s context, tone, and a number of other elements that come into play.

In most cases, if you choose to use an AI-written text, this choice itself means that you’ve taken a path of least effort, which is an impression you might not want to leave. In other words, it signals to the reader that the situation at hand is of minor importance, and that could go against your main business or personal goals.

In this sense, we’d recommend avoiding AI-written texts in the following occasions, but not limited to:

• An exchange of emails where the other person is actually writing. Simply pasting answers provided by an AI would most likely not be well received.
• Press releases. If the reader perceives that your PR piece was created by an AI, they could not take it as seriously as you wish them to.
• Feature announcements. When you create a new feature, you might want to give it the importance it deserves, and announce it with a proper specialist-written text.
• Journalistic content. AIs still don’t have – and we don’t know if they will – the accuracy and judgment for fact checking that journalists do.

Writing can be a demanding process, but there also lies the whole fun, beauty, and convincing power of it. However, mistake us not: AIs can be a positive and strong force within the writing field. Consider, for example, the possibilities they bring for ideation, grammar correction, translation, interpretation, and word choice.

Think about it: If you write an article yourself and then ask an AI to correct and revise it, the end result will be much different than if the article was created from a prompt in the first place. This differentiation is relevant to the topic of labeling. Content copied and pasted from AI should be clearly labeled as such.

The point we want to make is that we need to be selective when deciding on how AIs can complement our writing efforts, rather than fully giving in to the convenience that they provide.

Cheers to the conscious use of AI.