People want to live on the edge. Developers want to work with the latest cool technology, customers want to buy the new hotness. What's often forgotten is that most of those new cool technologies will be forgotten and discontinued in a few years. It's tricky to predict which ones will be around still after years.
You can always choose established technologies. But while there is more hope for them yo be supported for an extended period of time and the fact that they are uninspiring they too face the existential threat. Sometimes a new technology does indeed break through, often at the expense of some older technology making them obsolete.
We just finished a big upgrade today for one of our internal service. Several years ago it was built using an emerging technology stack that seemed promising at the time. While it's not our primary system it hasn't received much of any focus since it's creation. Why then did we have to upgrade it? Well, over the years the technology it was built on had evolved and the core version we used back then was recently deprecated rendering our service unmaintainable. It took several weeks just to port the core functionality over to the new version and there is still plenty of work to be done before everything will be up to date so we can get back to the original functionality without any added benefits (other than the service being usable once again).