Creative artificial intelligence can transform beliefs.

2023/06/22 Etxebeste Aduriz, Egoitz - Elhuyar Zientzia Iturria: Elhuyar aldizkaria

• Adimen artifiziala

Creative artificial intelligence systems can transform people's beliefs by transmitting false information and biases. Researchers Celeste Kidd and Abeba Birhane, in an analysis published in the journal Science, warn of the great capacity of systems such as ChatGPT, DALL-E or Midjorney.

Researchers have stressed that the capabilities of creative artificial intelligence have been exaggerated and have often been suggested to exceed human capabilities. In addition, these systems convey great security in the responses they give. And the human being tends to more and more rapidly internalize the information provided by an agent that seems expert and transmits security. Well, given that AI systems provide false, broad-filled information, researchers seriously see the risk of incorporating and spreading this kind of information into people's beliefs.

In addition, they have warned that once this information has been internalized and incorporated into beliefs, this could be difficult to change. Therefore, they stress that, while technology is still young, there is the possibility of interdisciplinary studies on this subject before being more deeply integrated into other everyday technologies.