AI Magazine May 2024 | Page 151

AI HARDWARE features like adaptive cruise control , lane-keeping assistance , and automated lane changes , enhancing driving safety and convenience .
Taking AI hardware to new heights , Tesla ’ s Full Self-Driving package incorporates the Full Self-Driving Computer ( FSD Computer ), a bespoke AI chip . Engineered specifically for autonomous driving , this potent hardware facilitates advanced tasks such as precise object detection , path planning , and decision-making , paving the way for fully autonomous vehicles .
Tesla MD Elon Musk recently shed light on the company ’ s substantial investment in AI hardware . In a post on X in January , Musk revealed that Tesla will spend more than US $ 500m on Nvidia hardware in 2024 . “ The table stakes for being competitive in AI are at least several billion dollars per year at this point ,” he said , highlighting the significance of AI hardware in maintaining competitiveness and driving innovation .
Amazon also recently announced it was using new chips for training and running AI . Amazon ’ s next-generation chips will be used for a wide range of cloud-based workloads and AI training models with the promise of better performance and energy efficiency .
One of the new chips is Trainium2 , meant for AI model training and said to deliver up to 4x better performance and 2x energy efficiency when compared to its predecessor . It is also expected to offer 3x more memory capacity than the first-gen Trainium chips .

“ Our GPUs are driving AI breakthroughs across industries . From medical imaging to selfdriving cars , AI hardware is revolutionising how we solve complex problems ”

JENSEN HUANG CEO , NVIDIA
David Brown , VP of Compute and Networking at Amazon Web Services says : “ With the surge of interest in generative AI , our chips will help customers train their ML models faster with better energy efficiency .”
Revolutionising AI hardware at Google and Microsoft Several other industry leaders are also leveraging AI hardware to revolutionise their respective domains . Google , for example , utilises TPUs extensively in its data centres to accelerate various AI workloads , enabling breakthroughs in natural language processing , image recognition , and more . Sundar Pichai , CEO of Google and Alphabet , states : “ TPUs have been transformative for
aimagazine . com 151