There's a paper (forgot the title or author, sorry) that gives the conclusion that it takes exponentially more resource to advance technologies forward in a lin

There's a paper (forgot the title or author, sorry) that gives the conclusion that it takes exponentially more resource to advance technologies forward in a linear fashion. Technological advances are not moving forward exponentially but rather it takes exponentially more resource to just keep progressing in linear scale. This applies to different disciplines across the board, in CS, in physics, chemistry, medical research, etc. It might relate to the productivity debate. It might take exponentially more advances in technologies to improve productivity linearly.