Optimising your hardware for your business’ data requirements: enabling unparalleled AI value

Meet the innovators taking AI to the next level of speed, efficiency and cost-effectiveness at the AI Hardware Summit, Mountain View, CA, September 18-19, 2018. - Ed Nelson - Conference Director, AI Hardware Summit

‘Business leaders need to understand the impact AI can have on their businesses, before their businesses are impacted. For example, they should map out how improvements in interaction with customers or suppliers could affect their business processes. Then, they should plan to adopt AI hardware and software solutions that can support these changes before they occur.’ – Chris Nicol, Co-Founder & CTO: Wave Computing.

Achieving or maintaining a competitive edge in today’s business world involves a comprehensive data strategy. In the realm of AI, where compute requirements and data volumes are dizzyingly high, it is especially crucial to evaluate which hardware needs to play a role in this strategy. As we move increasingly towards more heterogeneous computing models, it will be necessary to leverage different types of hardware for different AI workloads.

OpenAI reported in May that the ‘compute used in the largest AI training runs has been increasing exponentially with a 3.5 month-doubling time’ which is approximately 5 times faster than Moore’s Law. The report goes on to state that ‘improvements in compute have been a key component of AI progress, so as long as this trend continues, it’s worth preparing for the implications of systems far outside today’s capabilities.’ The appearance, according to Cade Metz of the New York Times, of 45 new AI chip start-ups, who have raised $1.5 billion in investment in 2017 alone, indicates that this trend is sure to continue.

2018 is the year of AI accelerators – in the past few months alone there have been AI chip announcements and rumours emanating across the world, from major semiconductor companies, CSPs, big tech companies and one or two automotive giants; plus developments and product releases in the start-up landscape.

Close to the entire global AI chip ecosystem, including Nicol, are gathering for the inaugural AI Hardware Summit at the Computer History Museum on September 18 – 19, 2018. This will be the first real opportunity to survey the architectural roadmap of AI accelerators. Presentations from the founders of the most exciting chip start-ups in the world, alongside technology leadership from CSPs and semiconductor companies, will cover technologies targeting both training and inference from the Cloud to the Edge. Explore HPC architectures, brain-inspired computing and quantum approaches to AI processing, alongside industry trends, investment and M&A.

We stand at the dawn of a new paradigm in computing – don’t get left behind!


‘It's important to optimize your hardware ecosystem to your business’ data requirements. You should understand how different types of hardware suit different needs – you certainly don’t want to be spending a ton of money on specialized hardware if you don’t have to - but to solve specific problems you may need AI accelerators.’ – Karl Freund, Senior Analyst, HPC & Machine Learning: Moor Insights & Strategy