When people talk about the benefits of artificial intelligence, the first argument they often hear is that it is ready for use 24/7 and never gets tired. But is that really the case? The so-called Neuromorphic Computing makes the crucial difference here. This AI is more similar to the human brain than has ever been the case before.
Neuromorphic computing ushers in the next generation of artificial intelligence. While the first generation of thinking machines was extremely rule-based and used pure logic to draw conclusions, this new type of artificial intelligence goes a step further. Pure reasoning based on specific rules is useful for monitoring processes and improving their efficiency. However, to bring real value in Artificial Intelligence, it needs to go beyond that.
Such is the case with Loihi 2. The artificial brain from Intel counts 120 million artificial synapses. The Neuromorphic Computing is based on a chip of the latest generation of the manufacturer and is particularly good at doing certain applications even more powerful and efficient than its predecessor Loihi 1 was.
Especially the processing of smells and sound recordings are better possible with the new AI than was previously the case. The AI learns from its mistakes using deep learning algorithms and is getting better and better at recognizing smells or sounds.
However, the disadvantage of chips is also not uncommon to humans. The AI can literally get tired. If the artificial intelligence of neuromorphic computing is used for too long, it needs breaks, otherwise its performance in recognition will suffer.
So humans and machines are becoming more and more similar. The AI learns from the behavior and brain of humans, both the positive characteristics and the weaknesses.
Post picture and video: Intel