Safety of self-assembled neuromorphic hardware
Neuromorphic devices are computational hardware inspired by our current understanding of the human brain. A multitude of designs has been proposed breaking with the conventional paradigm of binary transistors.
In the article, we assess whether advanced AI is more likely to emerge from nanowire devices than from conventional architectures. We focused on nanowire devices (shown in the gallery image).
Advanced AI is less likely to emerge from nanowire devices…
…due to high investments in standard architectures and open challenges in manufacturing nanowire devices. However, neuromorphic devices will be first choice for specialized edge devices due to low energy consumption.
Gallery image adapted from H. Sillin et al., “A theoretical and experimental study of neuromorphic atomic switch networks for reservoir computing,” Nanotechnology, vol. 24, p. 384004, Sep. 2013, doi: 10.1088/0957-4484/24/38/384004.