The Compute Unified Device Architecture (CUDA) was NVIDIA's GPGPU technology and even while OpenCL has come about as the industry standard, CUDA is still popular and is supported by NVIDIA hardware. NVIDIA also continues to advance CUDA with their latest compute architecture being found in the GeForce GTX 400 "Fermi" series hardware and their CUDA software on all operating systems continuing to advance.
Gregory Diamos, PhD student from Georgia Tech, along with his compiler research group though are now exploring the possibility of bringing up a CUDA back-end on Intel hardware within their open-source driver stack. "We are particularly interested in evaluating the Intel GPU architecture due to the tight integration between the CPU and GPU in GMA3150 and in the roadmaps for Sandybridge. After looking over the documentation in the programming guide, it seems like there is enough information of the ISA and buffer commands to do a backend for CUDA by generating an Intel binary for each CUDA program, and using buffer commands to setup memory, load the binary, and execute it."
This developer had a few questions about the Intel architecture and Linux driver, which were subsequently answered, and the discussion can be found on the intel-gfx mailing list. It's looking like a CUDA back-end for Intel hardware on Linux would be technically possible, but let's hope it ends up materializing.