The BrainChip AKD1500 SPI, announced on October 22, 2026, is a co-processor that brings Akida neuromorphic AI to wearables, smart home devices and other microcontroller-based products, without requiring manufacturers to redesign a product around a new processor.
The key point is the SPI interface, already present in most microcontrollers. The microcontroller keeps controlling the device and sends sensor data to the AKD1500 SPI, which runs the full neural network and returns the result, such as a recognized word or a detected fault.
The feature highlighted by the company is edge learning: the chip can learn new words, sounds, users or patterns directly on the device, without sending data to the cloud and without retraining the model. Like every Akida processor, it also runs several AI tasks on one chip.
Running the model on the device removes the round trip to the cloud, which cuts latency and keeps AI working even without connectivity, while voice, sensor and device-health data stay on the product. BrainChip cites robotics, industry and space-grade systems among the possible uses, and Onsor says the chip brings brain-inspired AI to its NEXA EEG Glasses, designed for everyday wear.
Alongside the AKD1500 PCIe, aimed at M.2 systems and multi-chip designs, the AKD1500 SPI opens the family to the high-volume microcontroller market, across consumer, industrial and healthcare products. According to BrainChip, the chip is already available from the company's shop, distributor DigiKey and direct sales.
As market context, the company cites an ABI Research forecast that TinyML chipset shipments, lightweight machine-learning models on ultra-low-power microcontrollers, will pass 4.1 billion units by 2031. That is a third-party forecast, not a result already achieved.
Fontes: BrainChip e GlobeNewswire.
Image credit: BrainChip AKD1500 chip, BrainChip press image via EIN Presswire. The photo shows the AKD1500 chip, from the same family as the new AKD1500 SPI.
Transparency: this article was produced with the help of artificial intelligence, cross-checked against public posts on X and internet sources. Check the original sources to confirm the details.
Muse
