PPubMed23 Dec 2020Cited 64×

Artificial Intelligence: A Clarification of Misconceptions, Myths and Desired Status Emmert-Streib et al. Artificial Intelligence

Frank Emmert-Streib, Olli Yli-Harja, Matthias Dehmer

Abstract

The field artificial intelligence (AI) was founded over 65 years ago. Starting with great hopes and ambitious goals the field progressed through various stages of popularity and has recently undergone a revival through the introduction of deep neural networks. Some problems of AI are that, so far, neither the “intelligence” nor the goals of AI are formally defined causing confusion when comparing AI to other fields. In this paper, we present a perspective on the desired and current status of AI in relation to machine learning and statistics and clarify common misconceptions and myths. Our discussion is intended to lift the veil of vagueness surrounding AI to reveal its true countenance.

§ The Valyu brief

The paper, in five minutes

Read from the full text via Valyu Contents

The paper clarifies AI misconceptions, noting lack of formal definition and that current AI methods address simple, high‑dimensional tasks like deep neural networks.

What’s new

i.Identifies the absence of a universally accepted definition of intelligence after 65 years of AI research
ii.Shows that AI’s original high‑level goals differ from its current data‑focused applications
iii.Demonstrates that deep neural networks, the most successful AI methods, address simple, high‑dimensional problems rather than complex, conceptual ones
iv.Clarifies that AI methods do not provide neurobiological models of the brain
v.Highlights the lack of a common mathematical formalism uniting AI methods

How they did it

The authors performed a literature‑based perspective analysis, reviewing historical AI definitions, common myths, and comparing AI methods with those from machine learning and statistics. They organized the discussion into sections on definitions, misconceptions, and method characteristics, and summarized findings in tables and figures.

What they found

No generally accepted formal definition of "intelligence" exists, even 65 years after AI’s inception
Deep neural networks, the most successful AI methods, focus on simple classification or regression tasks with high‑dimensional data
AI methods are fundamentally data‑analysis tools similar to those in machine learning and statistics
AI lacks a unified mathematical framework; its methods were introduced independently

Caveats

✕The universal intelligence measure relies on incomputable Kolmogorov complexity, limiting practical evaluation
✕Current AI methods do not provide biologically realistic models of brain function
✕Predictions about achieving strong AI remain vague and untestable
✕The paper is a conceptual perspective and does not include empirical experiments

§ The Valyu brief

Reading the full paper and taking notes. This takes a few seconds…

§ Ask this paper

Ask a question about this paper

Valyu reads the full text and answers from what the paper actually says.

§ Related work in other archives

Artificial Intelligence: A Clarification of Misconceptions, Myths and Desired Status Emmert-Streib et al. Artificial Intelligence

The paper, in five minutes

Ask a question about this paper

Similar papers in all four archives

Similar papers in all four archives

The paper, in five minutes