Programmable logic arrays have been around for a long time, they are basically a network of logic gates, where you burn them permanently into a useful configuration.
This is the same only a lot more complex. We are talking about configuring blocks into a whole AI capability, creating artificial synapses, burning it into something that can learn.
This is very much like taking a human brain with nothing in it (OK, no AOC jokes, this is serious), and training in a specialization, like language, or sight, or perhapts the other senses, like touch, hearing, smelling (processing from external sensors…).
This will be an amazingly transformative technology.
From the article: “Programmable resistors are the key building blocks in analog deep learning, just like transistors are the core elements for digital processors. By repeating arrays of programmable resistors in complex layers, researchers can create a network of analog artificial “neurons” and “synapses” that execute computations just like a digital neural network. This network can then be trained to achieve complex AI tasks like image recognition and natural language processing.”
New hardware offers faster computation for artificial intelligence, with much less energy
