The concept first arises in a paper entitled End-to-end arguments in system design by Jerome H. Saltzer, David P. Reed, and David D. Clark. They argue that reliable systems tend to require end-to-end processing to operate correctly, in addition to any processing in intermediate system. They then demonstrate that the end-to-end processing alone would suffice to make the system operate, and that the intermediate processing stages are largely redundant. Given this fact, much intermediate processing can be made simpler, relying on the end-to-end processing to make the system work. This leads to the model of a "dumb network" with smart terminals, a completely different model to the previous paradigm of the smart network with dumb terminals.
For example, in the TCP/IP protocol stack, IP is a dumb stateless protocol operating that simply moves datagrams across the network, and TCP is a smart end-to-end protocol operating between the client computers.
This paradigm was first made economically possible and then economically inevitable by the collapse in computer prices made possible by microprocessors.
References: