Advancement in artificial intelligence has led to the development of a new generation of specialized silicon chips that incorporate AI technology and are used for machine learning. Unlike the traditional chips, AI chips (also called AI hardware or AI accelerator) are specially designed for artificial neural network (ANN) based applications. They also incorporate huge smaller transistors, which run faster and consume less energy than larger transistors.
Today, there are scores of tech startups developing AI chips for different applications. One of the new entrants in the space is Untether AI, a Toronto, Canada-based startup that is developing ultra-efficient, high-performance AI chips to enable new frontiers in AI applications. The dramatic increase in the usage of AI, along with its heavy computational requirements, is overwhelming traditional compute architectures, and drastically increasing power consumption in data centers.
Untether AI’s at-memory compute architecture breaks through the computational bottleneck and changes the paradigm for AI compute efficiency. Untether AI’s tsunAImi accelerator cards powered by runAI devices provide record-breaking energy efficiency and compute density for inference acceleration.
By combining the power efficiency of at-memory computation with the robustness of digital processing, Untether AI has developed a groundbreaking new chip architecture for the neural net inference that eliminates the data movement bottleneck that costs energy and performance in traditional architectures.
To further accelerate the deployment of its at-memory compute technology across multiple markets, Untether announced it has raised a $125 million funding round co-led by Tracker Capital and Intel Capital, with participation from new investor Canada Pension Plan Investment Board (“CPP Investments”) and existing investor Radical Ventures. In conjunction with the funding, Untether also announced that Tracker Capital Senior Advisor Dr. Shaygan Kheradpir will join its AI’s Board of Directors.
Founded in 2018 by Darrick Wiebe, Martin Snelgrove, and Raymond Chik, Untether AI is creating a new class of specialized processors targeting an extreme point on the energy-complexity scale. By eliminating as much complexity as possible while still being able to perform the operations necessary to do neural net inference and putting that processing as close as possible to the memory, we can get close to theoretical minimum power numbers for the given process (initially targeting 28nm).
“I am pleased to add Tracker Capital to our prestigious group of investors and welcome Shaygan to our Board,” said Arun Iyengar, CEO, Untether AI. “Tracker Capital’s unmatched experience and relationships across sectors will help speed our engagements in multiple high-value markets, including telecom, technology, financial services, retail, and defense. I am also thrilled to welcome CPP Investments to the Untether AI family. With the new funding round and partnerships, we will be able to expand our current product reach and accelerate the development of our next-generation products.”
Dr. Kheradpir commented on the investment from Tracker Capital: “Untether AI has a scalable architecture that provides a revolutionary approach to AI inference acceleration. Its industry-leading power efficiency can deliver the compute density and flexibility required for current and future AI workloads in the cloud, for edge computing, and embedded devices.”
In addition to new investors Tracker Capital and CPP Investments, all existing investors participated in the round, which further validates the benefits of at-memory compute and the progress made by Untether AI.
Saf Yeboah-Amankwah, Intel Chief Strategy Officer added: “We have been an investor in Untether AI since the seed round. During that time Untether AI has assembled a world-class management team, developed and launched an exceptional product, and is now poised for growth in the burgeoning AI inference acceleration space.”
“AI has become a key enabling technology for many industries and Untether AI’s novel compute architecture has the potential to accelerate adoption across a number of use cases providing a strong value proposition to its customers,” said Leon Pedersen, Managing Director, Head of Thematic Investing, CPP Investments. “We seek to invest in innovative companies with strong market opportunities and we are pleased to support this leading Canadian AI technology company.”