What surprised me in the article was > “The goal of our Car Thing exploration in the U.S. was to learn more about how people listen in the car,” the company sa

What surprised me in the article was > “The goal of our Car Thing exploration in the U.S. was to learn more about how people listen in the car,” the company said in an emailed statement to TechCrunch. The goal was not to build a consumer device for users who actually wanted to use the damned thing! So it apparently was never about demand to begin with, it was a learning exercise. Apparently the RLHF was so disastrous it only took 5 months to train this feature out of Spotify's business plan.