From Von Neumann to AI Super Chips: The Story of AI Chips So Far
Moore’s law is no longer the key factor in chip development, AI is, and it’s transforming both chip design and the wider market in the process. GlobalData Thematic Research outlines the journey the industry has taken to reach the state of the AI chip market today, and what is set to come over the next decade
Von Neumann asserted that a general computer could simulate any effective procedure.
The phrase 'artificial intelligence' was first aired at a Dartmouth College conference.
Xilinx co-founders invented the first commercially viable FPGA.
Nvidia launched the GeForce 256, marketed as "the world's first GPU".
Amazon Web Services began providing an integrated suite of online services.
Google started testing robot cars on roads.
Apple launched the iPad, containing its own A4 chip. The A4 was also used in the iPhone 4.
Google DeepMind's AlphaGo algorithm beat the world Gochampion.
Microsoft re-engineered its server farms with FPGAs from Altera.
Google confirmed that it had been using the TPU in its data centers for more than a year.
Cambricon unveiled 1M, its third generation AI chip for edge devices.
The first appearance of Apple's A12 Bionic chip, featuring neural network hardware.
Huawei launched its Kirin 970 chip, including a neural network processing unit.
UK start-up Graphcore began shipping its machine learning chip, the Colossus GC2.
Alibaba revealed plans to produce its own AI chips, starting with the AliNPU.
Google Waymo launched a robot taxi service in Phoenix, Arizona.
Google's AlphaStar beat leading professional StarCraft II players.
Tesla introduced AutoPilot Hardware 3, powered by its own chip.
Amazon launches its Inferentia chip.
IBM commercialises its TrueNorth neuromorphic chip.
Intel brings its Loihi neuromorphic research chip to market.
Quantum computing reshapes supercomputing.
Deep learning becomes pervasive, powered by neuromorphic, quantum, optical, and organic computing.