Developing a new memory standard takes a long time, work on DDR4 memory for example started in 2005, a couple of years before the commercialization of DDR3 memory, so it doesn't come as a big surprise that work on High Bandwidth Memory (HBM) started a long time ago. KitGuru has written an interesting piece about how HBM came to be, and how this new industry standard was first conceived at ATI/AMD sometime around 2006-2007. HBM is expected to pick up the torch passed down by GDDR and may deliver video cards with 2TB/s or 4TB/s memory bandwidth in the next 5-7 years.
You can read the full piece over here.
By 2006 – 2007, when the work on the ATI R600 graphics processor with 512-bit memory bus as well as on the GDDR4 and the GDDR5 memory technologies was essentially completed, it became clear that memory consumed a lot of power already and would consume even more over time. Since ATI and Nvidia planned to use their GPUs for high-performance computing (HPC) applications, which require a lot of local memory, it was obvious that power consumption of GDDR was going to become a problem.
At the time, new memory tech development team at ATI Technologies led by Joe Macri came up with an idea of brand-new memory technology, which could provide extreme bandwidth while consuming a low amount of energy. The key elements of the new technology were multi-layer memory devices with an ultra-wide interfaces that used silicon interposer to connect to a processing device.