Imagine listening to a podcast that no one else has ever heard before and never will, you will be the only one that will ever hear what you are hearing. This is a possibility now with the creation of SHELDON by the University of California, Santa Cruz Ph.D. student James Ryan. SHELDON is the name of the artificial intelligence that hosts and creates the podcasts that you hear. The name is derived from the name Sheldon Klein, an early pioneer of expressive artificial intelligence.
Ryan's idea is that every person that tunes into the podcast will be listening to something unique. SHELDON works by generating counties which are called Sheldon Counties. The characters in these generated counties have their own lives with their own stories. They also interact with each other, which is what SHELDON turns into narratives which is what you hear if you tune into the podcast.
Although SHELDON hasn't been released yet, Ryan has uploaded a couple of samples of what some of the podcasts sound like. You can listen to them below.
The synthesized voice is creepy to say the least. Just thinking about the fact that it is not an actual person but in fact a computer that is pretending to be not only a human but an entire world of its own, there's just something not right about that.
While the samples posted on SoundCloud are the same for everyone, as they are just a recording, the final project is expected to be unique for every listener, creating a brand new world every single time it is accessed.
Ryan will complete his thesis regarding SHELDON this summer and is hoping to release a beta version for public listening in early 2019. Until the public release, Ryan has stated he will continue to release sample podcasts on both SoundCloud and Twitter.