A Step Towards Sustainable AI and Energy Efficiency

Last updated: 30 September 2024

In the ever-evolving landscape of artificial intelligence, energy consumption has become a critical bottleneck. With AI data centres consuming nearly 3% of global electricity, the urgency for energy-efficient and sustainable solutions cannot be overstated. As highlighted in a recent Forbes article, Literal Labs is tackling this head-on by pursuing a paradigm shift in AI through the Tsetlin Machine approach. We believe that the core challenge to AI’s future growth lies in addressing its energy demands—a challenge we are committed to solving.

The Tsetlin Machine approach marks a revolutionary departure from traditional neural networks, which are notoriously energy-hungry due to their reliance on matrix multiplications. By replacing these processes with a streamlined model utilising propositional logic, if-then statements, and voting algorithms, Literal Labs achieves a dramatic reduction in energy usage. Our solution delivers up to a thousandfold improvement in inferencing speed while preserving functionality, presenting what Forbes calls a “curve-jumping product.” This leap reimagines AI at its core rather than merely iterating on existing paradigms.

We are initially focusing our efforts on edge and IoT applications, where the energy efficiency of the Tsetlin Machine is most impactful. From anomaly detection, such as monitoring leaks and predictive maintenance, to applications in finance and insurance, the Tsetlin Machine’s efficiency and transparency provide transformative benefits. Unlike the opaque “black box” of neural networks, our linear, logical decision paths enable users to explain, understand, and ultimately trust AI outcomes, a feature in high demand across sectors seeking transparency.

Visit ‘Unlocking Sustainable AI: The Game-Changing Tsetlin Machine Approach’ to read the piece in full. It was authored by Charles Towers-Clark for Forbes and published 20 Sep 2024.