Damir explains at the ACP Unite 2022 conference in Vienna what he understands by AI. The functioning of the human brain, neuroscience, and the influence of neuroscientific research play a significant role in his explanation. The interconnections between IoT and the cloud are also vividly presented. The video explains how data is collected and trained in the cloud and how the concept of a Smart Factory works.
The development of AI is associated with high expectations, but also some uncertainties. What are the reasons for this, and what do you see as the opportunities?
Damir: “I must honestly admit that what is currently encompassed by the term ‘AI’ is not true artificial intelligence when examined closely. Today, we are primarily dealing with machine learning and statistics rather than intelligence. The ability to make a machine learn is fascinating and very practical. However, that does not make a machine intelligent or alive.
Some algorithms, such as neural networks, indeed utilize properties found in the brain. But all machine learning algorithms have one thing in common: they employ various mathematical procedures to solve specific problems.
The field I research is known as the Hierarchical Temporal Memory Cortical Learning Algorithm (HTM-CLA). It is a very new research area with the aim of reverse-engineering the neocortex, which is a part of the human cerebral cortex. Inspired by the functioning of the brain, the goal is to implement a so-called ‘cortical algorithm’ that can demonstrate true traces of intelligence. This algorithm does not rely on conventional mathematics; instead, it attempts to model the neocortex as it truly functions. Undoubtedly, this is a challenge, but perhaps not an unsolvable one. The initial results are promising.”
(Note: The translation provided above is a representation of the text in the given context. The explanation may not fully reflect the actual opinions or statements of Damir Dobric.)