Why invent AI systems that "understand, think, learn, and behave" like humans not possible?

Why machines, AI agents, neural nets, LLMs and GPT-n can not think or reason, learn and know like humans?

"Can AI think?” No.

“Can AI reason?” No.

"Can AI learn?" No.

"Can AI be hyper intelligent?" Yes.

Why machines, AI agents, neural nets, LLMs and GPT-n can not think or reason, learn and know like humans.

A wishful thinking or the widespread cases of global AI.

GPT-4 can solve difficult problems with greater accuracy, thanks to its broader general knowledge and problem solving abilities.

Gemini is built from the ground up for multimodality — reasoning seamlessly across text, images, video, audio, and code.

“What kind of mind does ChatGPT/Gemini have?”

“ChatGPT/Gemini can ace logic tests. But don’t ask it to be creative.”

“ChatGPT/ Gemini is dumber than you think”.

Human Thinking and Reasoning Generally, there are several main types of reasoning we employ when “thinking”: deduction and induction,

Abduction and analogy,

Statistical or probabilistic reasoning,

Inference and causal reasoning,

Plus automatic, unconscious,

Intuitive thinking,

If to reason is to think consciously and logically.

We think to reason with mental models to make a decision or solve a problem, to analyze, compare, and evaluate ideas or to infer something by reasoning from evidence.

Mental models help people make sense of the world—to interpret their environment and understand themselves. Mental models include categories, concepts, identities, prototypes, stereotypes, causal narratives, and worldviews. Without mental models of the world, it would be impossible for people to make most decisions in daily life.

Algorithmic Intelligence of Humans and Machine Human intelligence could be algorithmic as far as it involves systematic manipulations of abstract concepts, ideas, thoughts, images, numbers, signals, symbols or signs.

The concept of algorithm dated back to antiquity. Some arithmetic algorithms were used by ancient Babylonian mathematicians c. 2500 BC and Egyptian mathematicians c. 1550 BC.

All what we need, a complementary definition of human intelligence completing artificial intelligence as algorithmic intelligence classified by problems, their nature and complexity.

It is like a computational complexity of algorithms – the amount of time, storage, or other resources needed to execute them, all studied by the mathematical models of computation classified in three categories: sequential models, functional models, and concurrent models.

Broadly, algorithms are about optimal ways to solve a class of problems or to perform a computation, a strategy to reach a goal, "a set of rules that precisely defines a sequence of operations": specifications for performing thinking, reasoning or learning; instructions, steps and actions; computation and calculations, data processing, automated reasoning, or any other tasks.

Formally, it is an effective method, process, method, or procedure for decision problems to be expressed in terms of space and time, natural or formal languages, functions, inputs and outputs, or states, where the state transitions could be deterministic and probabilistic.

They are often divided by complexity, fields, implementation, design, optimization problem, or data types.

Some problems may have multiple algorithms of differing complexity, and all real problems have some algorithms or efficient algorithms, with mappings from some problems to other problems. So, the most suitable is to classify the problems themselves instead of the algorithms into equivalence classes based on the complexity of the best possible algorithms for them.

Generally, there are several kinds of problems-algorithms:

Philosophical algorithms as Philosophical methods;

Logical algorithms may be expressed as: Algorithm = logic + control;

Mathematical algorithms;

Mental algorithms coded in the brain;

Scientific algorithms as scientific methods;

Social, economic, political and governmental algorithms;

Computer algorithms as computer programs;

Automated reasoning in logical AI and expert systems’

Statistical classifiers or machine learning algorithms for statistical/stochastic/random/probabilistic processes;

AI algorithms as the master algorithms of all the above.

Today, computing algorithms of machine learning and deep learning are everywhere, from recommendation engines to legal algorithms, deciding human tastes, opinion, choices, rewards and punishment.

Machines, AI Agents, Neural Nets and LLMs can not Think or Reason like Humans, but compute and calculate, interpolate and extrapolate.

So, really real "AI is about COMPLETING, not simulating, human intelligence, organizing and leveraging the world's data/information/knowledge/intelligence making it universally accessible and useful".

As a fundamental general-purpose technology, it is integrating and leveraging emergent and digital technologies as special elements and modules:

Cyber-Physical AI = World Knowledge/Intelligence Models [World Model+ Foundation/Ontological Model + Scientific Models + Mathematical Models + Data Models + Language Models + Information Architecture + Global Knowledge Graph] +

RAI [Learning. Inference, Interaction] Engine +

Web Data + Web Intelligence +

Extended Reality [Augmented Reality (AR), Virtual Reality (VR), Mixed Reality (MR), and everything in between] + the Metaverse technology+

Unreal AI + ML + DL +

LLMs + Generative AI +

Causal/Explainable/Responsible/Trustworthy AI/ML/DL +

Interactive AI/ML/DL +

Cognitive Robotics + Intelligent Automation + Quantum Supercomputing +

Intelligent Cyber-Physical Systems + Internet of Things +

Emergent Technologies + Digital Intelligence Technologies + Climate Change Technologies + Space Technologies +

Intelligent Communities, Cities and Countries +

Hyper-Intelligent Hyper-Automation = the Intelligent Internet of Everything >

Global Digital Intelligence + Global Community Knowledge and Intelligence =

Global Man-Machine Hyperintelligence = Global Brain/Mind/Intellect Man-Machine Networks >

5I-World (Intelligent, Innovative, Interconnected, Instrumented, Inclusive).

Five main characteristics of Real AI are as follows.

World Knowledge and Intelligence: It will have access to massive amounts of world knowledge/data/information with the background knowledge on every domain, subject, theme and topic, problem and tasks.

Common Sense Knowledge and Reasoning: to develop common sense, from common sense physics to common sense psychology, to help AGI make intelligent decisions accordingly.

Transfer learning: It will be able to transfer the knowledge and skills of one task to other similar tasks.

Abstract Thinking/Conceptual Reasoning/Categorical Inference: It is able to understand and break down all the major ontological categories, abstract concepts, ideas, and thoughts.

Communicating in Computing Language or Natural Language, including NLP, NLG, NLU algorithms, as LLMs

Interaction/Cause and Effect: It is able to understand interactions of any scales, levels and complexities using cause and effect relationships to effectively interact with the world and its changing environments, situations and circumstances.

In reality, AI's true goal is not human intelligence, but reality, its truths and facts, models and simulations, all to effectively and sustainably interact with the world, learning and inferring, predicting and explaining all the predictable and unpredictable, as known and unknown outcomes, anticipated and unanticipated consequences, expected or unexpected behaviors.

#AItalent #AIresearch #AIdevelopment #AIadoption #AIhardware