Brace For Impact, Here Comes Personalized Ai

Intel, Apple, and Qualcomm are all racing to optimize their hardware and software to run artificial intelligence at the edge, meaning on local hardware, not remote cloud servers. This would allow for personalized, private AI that is so seamless it might be forgotten that it is AI. However, the most capable AI models today still rely heavily on expensive AI hardware and a reliable Internet connection. Professor Oliver Lemon of Heriot-Watt University believes that large language models are too slow to use for speech-based interaction.