# Are all these chips really Brain Inspired? // The AI Hardware Show S2E7 **URL:** \ **Category:** Content Creators\ **Tags:** hardware, techtechpotato\ **Created:** [22 June 2026 12:00 UTC](https://www.artofsm.art/t/are-all-these-chips-really-brain-inspired-the-ai-hardware-show-s2e7/20432 "2026-06-22T12:00:29Z")\ **Posts on this page:** 2\ **Page:** 1
**Author:** ![merefield](https://www.artofsm.art/user_avatar/www.artofsm.art/merefield/32/74_2.png) [@merefield](https://www.artofsm.art/u/merefield)\ **Post date:** [22 June 2026 12:00 UTC](https://www.artofsm.art/t/are-all-these-chips-really-brain-inspired-the-ai-hardware-show-s2e7/20432/1 "2026-06-22T12:00:29Z")
[![](https://www.artofsm.art/uploads/default/original/3X/b/0/b02e67070b7ebdd0f0f1b62e981f15a980d37d90.jpeg "Are all these chips really Brain Inspired? // The AI Hardware Show S2E7") ](https://www.youtube.com/watch?v=EZrkVy5WXXk) The episode reviews several neuromorphic chips, including Brainchip’s Akida, Intel’s Loihi 2, Sincense’s Spec, Innotera’s Pulsar, and SpinCloud’s Spica 2, highlighting their brain-inspired designs, low-power consumption, and specialized applications in edge AI, robotics, and time series analysis. While each chip employs different architectures—spiking neural networks, analog neurons, or massively parallel digital cores—they collectively advance practical, energy-efficient AI hardware beyond traditional models, targeting real-time, event-driven tasks rather than large-scale deep learning. ---
**Author:** ![merefield](https://www.artofsm.art/user_avatar/www.artofsm.art/merefield/32/74_2.png) [@merefield](https://www.artofsm.art/u/merefield)\ **Post date:** [22 June 2026 12:21 UTC](https://www.artofsm.art/t/are-all-these-chips-really-brain-inspired-the-ai-hardware-show-s2e7/20432/3 "2026-06-22T12:21:00Z")
The episode of the AI Hardware Show explores various neuromorphic chips, highlighting their brain-inspired designs and practical applications. Brainchip’s Akida chip is a commercial spiking neural network accelerator designed for ultra-low power AI at the edge. It mimics biological neurons by processing discrete electrical spikes, capturing both spike timing and magnitude for richer information encoding. Akida supports one-shot learning, enabling real-time recognition tasks like voice control and face recognition without cloud dependency. Its developer-friendly software integrates with popular AI frameworks, and the second generation even supports transformer models, expanding its use cases beyond simple sensor workloads. Intel’s Loihi 2 chip represents a fully digital, asynchronous neuromorphic processor built on advanced Intel 4 process technology. It features a mesh of spiking neural cores capable of supporting one million neurons and 120 million synapses, focusing on low-power, real-time decision-making for edge applications. Intel complements the hardware with the open-source Lava software stack, which facilitates development across neuromorphic platforms and fosters academic and industrial collaborations. While not suited for large-scale models like GPT, Loihi 2 excels in sparse, event-driven AI tasks such as robotics and adaptive control, positioning it as a pioneering experiment in brain-inspired silicon. Sincense’s Spec chip combines a proprietary neuromorphic processor with a dynamic vision sensor that reacts only to changes in brightness, drastically reducing power consumption and latency. This fully asynchronous architecture eliminates traditional clock signals, achieving milliwatt-level power usage and sub-5 millisecond response times. Targeting edge applications like gesture recognition and autonomous driving assistance, Spec emphasizes practical deployment with a clear product roadmap. The company has secured significant funding to develop next-generation 3D neuromorphic processors, aiming to scale compute density while maintaining ultra-low power consumption. Innotera’s Pulsar chip takes a unique analog approach, integrating analog neurons that fire voltage spikes to process time series data efficiently. This mixed-signal architecture excels in applications such as keyword spotting and heartbeat analysis, consuming energy levels close to biological neurons. Unlike many neuromorphic efforts focused on vision, Pulsar targets domains where analog spike timing offers distinct advantages over digital accelerators. Its T1 silicon implementation combines analog spiking neural fabrics with digital processing cores, reflecting a pragmatic balance between biological efficiency and programmability. The company is focused on market education to highlight the importance of time series data in AI. SpinCloud, originating from the University of Manchester’s Spica project, offers a massively parallel digital neuromorphic system designed for real-time brain-like simulations. Its Spica 2 chip integrates 152 ARM-based cores optimized for event-driven workloads, with large-scale deployments supporting millions of cores and billions of neurons. Unlike startups selling chips, SpinCloud builds full infrastructure platforms targeting complex, sparse, and hybrid AI models with ultra-low latency. Funded primarily through EU grants, the company aims to lead the “third wave” of AI by delivering scalable, energy-efficient neuromorphic systems for applications ranging from national defense to smart cities.