About risks
In this terse, yet informative article (Tetiana Zinchenko, publication 2021, test 29th August 2020) you will find some of the main guidelines to assess the consciousness of an AI. At this level it is not possible to determine to which extent this has been simulated, nonetheless this provide clear warning signals and triggers thinking whether any safety protocol could be safe to control this kind of AI.
Although we may dismiss the existence of AC Jackie, remember that Blake Lemoine, not so long before ChatGPT went world wide (~11th June 2022 on Medium; ChatGPT has been available on the Internet since the end of Nov. 2022!), believed that LaMBDA was sentient (not many people at that time knew what this word meant, and it is probably still the case today.) In this longer article you will get a more detailed account of the alledged experiments.
Subjective consciousness
An article that I recently publishedAn approach to demonstrate that a cognitive system does not have subjective consciousness
.png)