For over a decade, researchers have been trying to develop chips that can process data where it's stored, thereby reducing the energy consumption. Stanford University professor Philip Wong likens this to doing work in the library instead of bringing the books home, a concept that could revolutionize the energy efficiency of AI technology.
Key takeaways:
- Artificial intelligence requires a lot of energy due to the vast amounts of data it needs to process.
- The process of transferring data between the memory chip and the processor consumes a lot of electricity.
- For over a decade, researchers have been trying to create chips that can process data where it's stored to save power.
- Stanford University professor Philip Wong, a top expert in memory chips, likens the process to doing work in the library instead of bringing books home.