Artificial intelligence is penetrating every aspect of our lives and work, and it’s constantly becoming more human-like. Whether it’s ChatGPT’s advanced voice feature that has a warm, engaging, confident, and charismatic voice or Character.ai that lets you talk to famous figures like Elon Musk or even Napoleon Bonaparte, AI is becoming more trustworthy. But does making AI more human-like make it more trustworthy? The answer, researchers say, is no.

It is a very interesting question whether the more human-like an AI is, the more trustworthy it is. While some may assume that human-likeness can instill more trust, research shows that this is not always the case.

One reason for this may be that when AI becomes too human-like, it can create feelings of insecurity or skepticism in users. The term “Uncanny Valley” refers to this phenomenon where a close, but not entirely realistic, resemblance to humans can increase feelings of insecurity and discomfort.

Trust in AI is based more on its performance, transparency, and ethics. It is important for users to understand how the AI ​​makes decisions and to be able to trust the information and results it provides.