Intelligence in Artificial Intelligence

Shoumen Palit Austin Datta

DOI: https://doi.org/10.48550/arXiv.1610.07862

2016-10-26

Abstract:The elusive quest for intelligence in artificial intelligence prompts us to consider that instituting human-level intelligence in systems may be (still) in the realm of utopia. In about a quarter century, we have witnessed the winter of AI (1990) being transformed and transported to the zenith of tabloid fodder about AI (2015). The discussion at hand is about the elements that constitute the canonical idea of intelligence. The delivery of intelligence as a pay-per-use-service, popping out of an app or from a shrink-wrapped software defined point solution, is in contrast to the bio-inspired view of intelligence as an outcome, perhaps formed from a tapestry of events, cross-pollinated by instances, each with its own microcosm of experiences and learning, which may not be discrete all-or-none functions but continuous, over space and time. The enterprise world may not require, aspire or desire such an engaged solution to improve its services for enabling digital transformation through the deployment of digital twins, for example. One might ask whether the "work-flow on steroids" version of decision support may suffice for intelligence? Are we harking back to the era of rule based expert systems? The image conjured by the publicity machines offers deep solutions with human-level AI and preposterous claims about capturing the "brain in a box" by 2020. Even emulating insects may be difficult in terms of real progress. Perhaps we can try to focus on worms (Caenorhabditis elegans) which may be better suited for what business needs to quench its thirst for so-called intelligence in AI.

Artificial Intelligence

What problem does this paper attempt to address?

This paper attempts to explore the essence of "intelligence" in artificial intelligence (AI), as well as the possibilities and challenges of its implementation. Specifically, the paper focuses on the following aspects:

1. **Definition of intelligence**: The paper discusses the basic elements that constitute intelligence and points out that intelligence is not a simple point, but a continuous process or a collection of network systems. This view emphasizes the complexity and multi - dimensional characteristics of intelligence.

2. **Historical review**: The paper reviews the journey of artificial intelligence from the "winter" in the 1990s to its peak in 2015, and points out the different understandings and expectations of intelligence during this process.

3. **Biological inspiration**: The paper explores the inspiration of biology for intelligence research, especially by studying simple organisms (such as the nematode *Caenorhabditis elegans*) to understand the formation mechanism of intelligence. This shows that even in simple organisms, intelligence can be complex, dynamic, and affected by multiple factors.

4. **Technical and commercial applications**: The paper analyzes the current situation of AI technology in commercial applications, especially the concept of "Intelligence as a Service". The author has reservations about this commercialization trend, believing that these technologies may not have truly achieved human - level intelligence.

5. **Gap between theory and practice**: The paper discusses the gap between theoretical models and practical applications, especially in the fields of neural networks and machine learning. The author points out that existing AI systems still have limitations when dealing with complex tasks, especially in simulating human cognitive and learning abilities.

6. **Future prospects**: Finally, the paper presents the prospects for future research, emphasizing the importance of continuing to explore and understand intelligence, although the current technology is far from achieving human - level intelligence.

Overall, this paper aims to explore the essence of "intelligence" in artificial intelligence and, through the analysis of multiple perspectives such as history, biology, technology, and commerce, reveals the challenges and future possibilities of achieving true intelligence.