BrainChip gets a mention, along with Syntiant, just before the 10 min mark.
[08:22] Nice, we don't need massive TOPS or heavy headpipes ... we're on the side which is more cost-effective, power-efficient, and easier to program, making them a better choice for applications with simpler computational requirements and power constraints.
The choice between an FPGA and an MCU depends on the specific requirements of the application, such as performance, power consumption, flexibility, cost, and development effort.
That's why we target edge devices and beeing not an NVIDIA rival... because, in some cases, a hybrid approach combining both technologies may be the optimal solution.
Cost is often the dominant factor in why a microcontroller or an ASIC is chosen over an FPGA for large-volume applications.