Why today’s AI is not AI, ML is not ML and DL is not DL, or how to build an authentic, trustworthy, real-world AI

Machine Learning with Deep Learning AI is today’s mainstream AI, and it is promising to become omnipresent and omniscient with virtually unlimited applications in all parts of everyday life.

Still today’s machine learning AI solutions are not intelligent in the ways natural human intelligence demonstrates as well as popularized in sci-fi literature and Hollywood movies depicting scores of self-knowing machines, sensitive robots, sapient androids and gynoids.

In all, great progress has been made in improving machine’s power to compute very complex problems, situations, and events. However, real deep intelligence with its true recognition, comprehension and abstraction, learning, knowing and self-knowing, reasoning and understanding, remains beyond the reach of ML/AI today.

The ML/AI solutions are powerful but with principal limitations. ML/AI it will adapt but not create, talk but not walk, specialize but not generalize, fail to discriminate between things, as man and machine, automate, augment, or extend but not originate, communicate but not understand. It has no causal powers of seeing, smelling, hearing, speaking and writing, feeling, and recognizing people, objects and scenes, and especially thinking, as to generate, analyze and integrate knowledge from diverse information inputs and signals, such as environmental conditions, biometrics or psychographics.  [Making Sense of AI, SAS, 2021]

The principal limitations of today’s ML/DL are in its inherent incapacity to make sense of the world of reality. For the real and true AI is about modeling reality with its physicality and mentality in digital reality by intelligent computing machinery as complementing human intelligence.

The principal limitation of today’s mainstream ML/DL/AI is to be defined, designed, and developed following a non-scientific anthropomorphic and anthropocentric paradigm, simulating human brain, intelligence, mind or actions.     

This all gives a foundation to argue that today’s AI is not AI as well as ML is not ML and DL is not DL and the whole great enterprise is in the urgent need of the AI/ML/DL paradigm shift. Being specialized, weak and narrow systems, biased and judgemental, not really intelligent, means not being autonomous, self-knowing, interactive, dynamic, causal, general, intuitive, creative, productive, smart, humane, ethical, lawful, responsible, or trustworthy.

Most existential concerns come from the ethical issues of the mainstream AI: unemployment, inequality, artificial stupidity, humanity, AI bias or racists robots, security, evil genies, singularity, robot rights. https://www.weforum.org/agenda/2016/10/top-10-ethical-issues-in-artificial-intelligence/

In turn, according to the EU Ethics Guidelines, trustworthy AI should be:

(1) lawful -  respecting all applicable laws and regulations

(2) ethical - respecting ethical principles and values

(3) robust - both from a technical perspective while taking into account its social environment

https://digital-strategy.ec.europa.eu/en/library/ethics-guidelines-trustworthy-ai

 

Everything starts from your assumptions and definitions or paradigms

As such, the mainstream AI/ML/DL has some advantages, but many disadvantages, and everything will depend on how AI will be interpreted or mis-interpreted, used or misused.

These contradictions come from its subjective mainstream definition, as “simulating the human intelligence processes in machines, as computing systems” or “the hardware and software mimicking human the functions of the human brain, mind, behaviour or reasoning to solve complex problems” (Cognitive Computing). 

The first European Commission's definition followed the suit: “Artificial intelligence (AI) refers to systems that display intelligent behaviour by analysing their environment and taking actions – with some degree of autonomy – to achieve specific goals.

But the updated definition of AI is closer to the Real AI model: “Artificial intelligence (AI) refers to systems designed by humans that, given a complex goal, act in the physical or digital world by perceiving their environment, interpreting the collected structured or unstructured data, reasoning on the knowledge derived from this data and deciding the best action(s) to take (according to pre-defined parameters) to achieve the given goal. AI systems can also be designed to learn to adapt their behaviour by analysing how the environment is affected by their previous actions”. https://ec.europa.eu/futurium/en/system/files/ged/ai_hleg_definition_of…

As far as our rate of technological advancement increases, 70-years-old AI/ML/DL needs its paradigm shift, a major change in its philosophy, concepts, and practices of how AI works or is accomplished. Paradigms are important both for humans and machines because they define how intelligent agents perceive reality and how they behave within it.

We have to found all our study and practice on the solid scientific definition:

“AI is the master computing algorithm, code or program that learns, understands, reasons and effectively interacts with any environments, physical, mental, social, digital, or virtual”.

A paradigm shift results after the accumulation of anomalies or evidence, as inconsistent advantages and disadvantages, that challenges the status quo, or due to some revolutionary innovation or discovery.

The paradigm shift implies a major change in how people think and get things done, upending and replacing the prior paradigms, as of symbolic AI or statistic ML.

The AI paradigm shift consists in the disruptive idea of Transdisciplinary Human-Machine Superintelligence [Supermind] implying the Deep Causal AI or Real-World AI model for Narrow AI, ML, DL and Human Intelligence.

The Deep AI internet will be disruptively groundbreaking, creating a paradigm shift in how people live, think, work, study, and play and get their information, communicate, and shop. The first I-commerce companies to capitalize on this new paradigm will thrive, living many internet firms as well as the big tech companies, as Amazon, Apple, Google, Facebook, Microsoft, went out of business.

For those who occupy the established paradigm are traditionally reluctant, or even hostile, toward contradictory theories or evidence that challenge their worldview or practices.

Besides, the Deep AI paradigm shift is a radical change in the perception of how things should be thought about, done, or made. It can require educational programs to be reviewed, national AI strategies to be recalled, entire institutions, companies, and departments to be eliminated or created. Trillions of dollars of new deep ML/AI platforms need to be deployed while the old cloud AI platforms and social media networks applications with their internet search engines are to be disrupted or renovated, as listed below:

Top 10 Social Networking Sites by Market Share Statistics [2021]

  • Facebook – 2.74 Billion Active Users.
  • YouTube – 2.291 Billion Active Users.
  • WhatsApp – 2.0 Billion Active Users.
  • Facebook Messenger – 1.3 Billion Active Users.
  • Instagram – 1.221 Billion Active Users.
  • Weixin/WeChat – 1.213 Billion Active Users.

https://www.linkedin.com/pulse/trans-ai-eat-world-azamat-abdoullaev/?published=t

Will Artificial Superintelligence (ASI) Create Infinite Economic Growth or More Inequality?

Machine Learning with Deep Learning is just advanced Data Analytics techniques

Machine Learning with Deep Learning is just advanced Data Analytics techniques. Both terms, ML and DL, are harmfully misleading and overhyped, diverting us from the real AI study and development.

ML is just one of many statistical techniques in an advanced Data Analytics, analyzing current and historical data/facts to make predictions about future or past or otherwise unknown events.

Here is a rather honest recognition from SAS: Machine learning automates analytical model building. It uses methods from neural networks, statistics, operations research and physics to find hidden insights in data without explicitly being programmed for where to look or what to conclude.

What are the types of machine learning? What are the process involved while applying machine learning? Where to practice for Machine learning and the maths and statistics behind machine learning. How machine learning and deep learning is related to data science. All is sketched in the diagram below. https://t.me/dataspoof

 

Legally speaking, all the organizations which impersonate it as an AI are involved in a sort of crime, doing AI washery and marketing on massive scale false services and products, or fake AIs.

But to prosecute all of them, you need a standard and criterion, which is a real and true AI, mostly unwelcome by the big tech, as G-MAFIA or BAT-triada.

Suppose, I have created a real-world AI model with all the key intelligent functionalities, then a team of lawyers could action all the fake AI underworld for Euro trillions compensations for massive fraud and illegal profiting.

https://futurium.ec.europa.eu/en/european-ai-alliance/posts/man-machine-superintelligence-paradigm-shift-trans-ai-deep-ai-or-real-world-ai-integral-trustworthy

It is all data analytics….

To the point. Don’t mix, data analysis with data analytics and data science.

Data analysis is the science of analyzing/inspecting/cleansing/transforming/modeling raw data discovering useful information, informing conclusions, and supporting decision-making.

  • The techniques and processes of data analytics have been automated into mechanical processes and algorithms that work over raw data for human consumption.

Dara Analytics is a broader term embracing diverse types of data analysis. It is generally defined as “the systematic computational analysis of data or statistics”. It is used for the discovery, interpretation, and communication of meaningful patterns in data, applying data patterns to decision-making.

Data Analytics areas include machine learning, predictive analytics, prescriptive analytics, enterprise decision management, descriptive analytics, cognitive analytics, Big Data Analytics, retail analytics, supply chain analytics, store assortment and stock-keeping unit optimization, marketing optimization and marketing mix modeling, web analytics, call analytics, speech analytics, sales force sizing and optimization, price and promotion modeling, predictive science, graph analytics, credit risk analysis, and fraud analytics.

General types of Data Analytics

  1. Descriptive analytics: This describes what has happened over a given period of time.
  2. Diagnostic analytics: This focuses more on why something happened.
  3. Predictive analytics: This moves to what is likely going to happen in the near term.
  4. Prescriptive analytics: This suggests a course of action.

It is long used in marketing, credit risk assessment, fraud detection, manufacturing, healthcare, and government operations including law enforcement.

Data predictive analytics wiki article gives a long listing where it is used:

actuarial science, marketing, business management, sports/fantasy sports, insurance, policing, telecommunications, retail, travel, mobility, healthcare, child protection, pharmaceuticals, capacity planning, social networking and other fields.

Data science goes as an interdisciplinary field that uses scientific methods, processes, algorithms and systems to extract knowledge and insights from data, structured and unstructured. It is related to data mining, machine learning and big data, using techniques and theories from many mathematics, statistics, computer science, information science, and domain knowledge.

Data science is "to unify statistics, data analysis, informatics, and their related methods" to "understand and analyze actual phenomena" with data. It is pushed as a "fourth paradigm" of science, additionally to theoretical empirical and computational.

Where is the whole trick?

The trick is data analytics has been adopted by many sectors. Big Data analytics can require extensive computation, the algorithms and software used for BD analytics harness the most current methods in computer science, statistics, and mathematics. Just dubbed it under the most sexy name of AI, and sell it with any profit.

The travel and hospitality industry collect customer data and figure out where the issues lie and how to fix them.

Healthcare combines the use of high volumes of structured and unstructured data and uses data analytics to make quick decisions. The retail industry uses copious amounts of data to meet the ever-changing demands of shoppers to identify trends, recommend products, and increase profits.

The big tech companies and social networks services, as Facebook, and content companies use the same data analytics techniques to keep you clicking, watching, or re-organizing content to get another view or another click.

Gaming companies use data analytics to set reward schedules for players that keep the majority of players active in the game, etc.

So, when you are told that you enjoy some AI/ML/DL products or services keep in mind the following. It is a false advertisement of fake AI. Such an untrue or misleading information given to you to get you to buy something, or to come visit their site, platform or store. Those who make and sell fake ML/AI products must honestly present their products, services and prices to you.

For example, the IBM team sells its Summit supercomputer as a new generation of smart AI supercomputers, which has ZERO intelligence, no awareness of what it is doing and computing.

“Summit, an IBM AC922 system, links more than 27,000 NVIDIA Volta GPUs with more than 9,000 IBM Power9 CPUs to provide unprecedented opportunities for the integration of AI and scientific discovery.

Applying AI techniques like deep learning to automate, accelerate, and drive understanding at supercomputer scales will help scientists achieve breakthroughs in human health, energy, and engineering and answer fundamental questions about the universe.

The arrival of AI supercomputing and Summit means science has never been smarter”.

https://www.olcf.ornl.gov/summit/

Legally, you can’t blame the AI underworld in selling the fake ML/DL/AI product as far as the real-world AI with its definition, standards, specifications and technical parameters is formally adopted.

Sources

https://futurium.ec.europa.eu/en/user/10596

Critical Commentary on the “High-Level Conference on AI: From ambition to Action” Program

AI is in the critical stage of its revolutionary paradigm shift, as Transdisciplinary Science and Technology and Real-World Man-Machine Intelligence,…

Man-Machine Superintelligence Paradigm Shift: Trans-AI, Deep AI, or Real-World AI, as an integral Trustworthy AI Model

AI is becoming omnipresent and omniscient with virtually unlimited applications in all parts of everyday life. As such, it has many advantages,…

Trans-Technologies: AI as Transdisciplinary Science and Technology: how to solve the biggest challenges in the AI Industry

AI is set to change the ways the future world to work. What kind of future AI technology is to domineer will decide the kind of future world to live…

The Trans-AI will eat the world

Apple, Google, Amazon, Microsoft, and Facebook, enjoying the highest ever $1+ trillion market cap, are still working on an old and…

Machine Learning with Deep Learning AI is today’s mainstream AI and it is promising to become omnipresent and omniscient with virtually unlimited applications in all parts of everyday life.

Still today’s machine learning AI solutions are not intelligent in the ways natural human intelligence demonstrates as well as popularized in sci-fi literature and Hollywood movies depicting scores of self-knowing machines, sensitive robots, sapient androids and gynoids.

In all, great progress has been made in improving machine’s power to compute very complex problems, situations, and events. However, real deep intelligence with its true recognition, comprehension and abstraction, learning, knowing and self-knowing, reasoning and understanding, remains beyond the reach of ML/AI today.

The ML/AI solutions are powerful but with principal limitations. ML/AI it will adapt but not create, talk but not walk, specialize but not generalize, fail to discriminate between things, as man and machine, automate, augment, or extend but not originate, communicate but not understand. It has no causal powers of seeing, smelling, hearing, speaking and writing, feeling, and recognizing people, objects and scenes, and especially thinking, as to generate, analyze and integrate knowledge from diverse information inputs and signals, such as environmental conditions, biometrics or psychographics.  [Making Sense of AI, SAS, 2021]

The principal limitations of today’s ML/DL are in its inherent incapacity to make sense of the world of reality. For the real and true AI is about modeling reality with its physicality and mentality in digital reality by intelligent computing machinery as complementing human intelligence.

The principal limitation of today’s mainstream ML/DL/AI is to be defined, designed, and developed following a non-scientific anthropomorphic and anthropocentric paradigm, simulating human brain, intelligence, mind or actions.     

This all gives a foundation to argue that today’s AI is not AI as well as ML is not ML and DL is not DL and the whole great enterprise is in the urgent need of the AI/ML/DL paradigm shift. Being specialized, weak and narrow systems, biased and judgemental, not really intelligent, means not being autonomous, self-knowing, interactive, dynamic, causal, general, intuitive, creative, productive, smart, humane, ethical, lawful, responsible, or trustworthy.

Most existential concerns come from the ethical issues of the mainstream AI: unemployment, inequality, artificial stupidity, humanity, AI bias or racists robots, security, evil genies, singularity, robot rights. https://www.weforum.org/agenda/2016/10/top-10-ethical-issues-in-artificial-intelligence/

In turn, according to the EU Ethics Guidelines, trustworthy AI should be:

(1) lawful -  respecting all applicable laws and regulations

(2) ethical - respecting ethical principles and values

(3) robust - both from a technical perspective while taking into account its social environment

https://digital-strategy.ec.europa.eu/en/library/ethics-guidelines-trustworthy-ai

 

Everything starts from your assumptions and definitions or paradigms

As such, the mainstream AI/ML/DL has some advantages, but many disadvantages, and everything will depend on how AI will be interpreted or mis-interpreted, used or misused.

These contradictions come from its subjective mainstream definition, as “simulating the human intelligence processes in machines, as computing systems” or “the hardware and software mimicking human the functions of the human brain, mind, behaviour or reasoning to solve complex problems” (Cognitive Computing). 

The first European Commission's definition followed the suit: “Artificial intelligence (AI) refers to systems that display intelligent behaviour by analysing their environment and taking actions – with some degree of autonomy – to achieve specific goals.

But the updated definition of AI is closer to the Real AI model: “Artificial intelligence (AI) refers to systems designed by humans that, given a complex goal, act in the physical or digital world by perceiving their environment, interpreting the collected structured or unstructured data, reasoning on the knowledge derived from this data and deciding the best action(s) to take (according to pre-defined parameters) to achieve the given goal. AI systems can also be designed to learn to adapt their behaviour by analysing how the environment is affected by their previous actions”. https://ec.europa.eu/futurium/en/system/files/ged/ai_hleg_definition_of…

As far as our rate of technological advancement increases, 70-years-old AI/ML/DL needs its paradigm shift, a major change in its philosophy, concepts, and practices of how AI works or is accomplished. Paradigms are important both for humans and machines because they define how intelligent agents perceive reality and how they behave within it.

We have to found all our study and practice on the solid scientific definition:

“AI is the master computing algorithm, code or program that learns, understands, reasons and effectively interacts with any environments, physical, mental, social, digital, or virtual”.

A paradigm shift results after the accumulation of anomalies or evidence, as inconsistent advantages and disadvantages, that challenges the status quo, or due to some revolutionary innovation or discovery.

The paradigm shift implies a major change in how people think and get things done, upending and replacing the prior paradigms, as of symbolic AI or statistic ML.

The AI paradigm shift consists in the disruptive idea of Transdisciplinary Human-Machine Superintelligence [Supermind] implying the Deep Causal AI or Real-World AI model for Narrow AI, ML, DL and Human Intelligence.

The Deep AI internet will be disruptively groundbreaking, creating a paradigm shift in how people live, think, work, study, and play and get their information, communicate, and shop. The first I-commerce companies to capitalize on this new paradigm will thrive, living many internet firms as well as the big tech companies, as Amazon, Apple, Google, Facebook, Microsoft, went out of business.

For those who occupy the established paradigm are traditionally reluctant, or even hostile, toward contradictory theories or evidence that challenge their worldview or practices.

Besides, the Deep AI paradigm shift is a radical change in the perception of how things should be thought about, done, or made. It can require educational programs to be reviewed, national AI strategies to be recalled, entire institutions, companies, and departments to be eliminated or created. Trillions of dollars of new deep ML/AI platforms need to be deployed while the old cloud AI platforms and social media networks applications with their internet search engines are to be disrupted or renovated, as listed below:

Top 10 Social Networking Sites by Market Share Statistics [2021]

  • Facebook – 2.74 Billion Active Users.
  • YouTube – 2.291 Billion Active Users.
  • WhatsApp – 2.0 Billion Active Users.
  • Facebook Messenger – 1.3 Billion Active Users.
  • Instagram – 1.221 Billion Active Users.
  • Weixin/WeChat – 1.213 Billion Active Users.

https://www.linkedin.com/pulse/trans-ai-eat-world-azamat-abdoullaev/?published=t

Will Artificial Superintelligence (ASI) Create Infinite Economic Growth or More Inequality?

Machine Learning with Deep Learning is just advanced Data Analytics techniques

Machine Learning with Deep Learning is just advanced Data Analytics techniques. Both terms, ML and DL, are harmfully misleading and overhyped, diverting us from the real AI study and development.

ML is just one of many statistical techniques in an advanced Data Analytics, analyzing current and historical data/facts to make predictions about future or past or otherwise unknown events.

Here is a rather honest recognition from SAS: Machine learning automates analytical model building. It uses methods from neural networks, statistics, operations research and physics to find hidden insights in data without explicitly being programmed for where to look or what to conclude.

What are the types of machine learning? What are the process involved while applying machine learning? Where to practice for Machine learning and the maths and statistics behind machine learning. How machine learning and deep learning is related to data science. All is sketched in the diagram below. https://t.me/dataspoof

 

Legally speaking, all the organizations which impersonate it as an AI are involved in a sort of crime, doing AI washery and marketing on massive scale false services and products, or fake AIs.

But to prosecute all of them, you need a standard and criterion, which is a real and true AI, mostly unwelcome by the big tech, as G-MAFIA or BAT-triada.

Suppose, I have created a real-world AI model with all the key intelligent functionalities, then a team of lawyers could action all the fake AI underworld for Euro trillions compensations for massive fraud and illegal profiting.

https://futurium.ec.europa.eu/en/european-ai-alliance/posts/man-machine-superintelligence-paradigm-shift-trans-ai-deep-ai-or-real-world-ai-integral-trustworthy

It is all data analytics….

To the point. Don’t mix, data analysis with data analytics and data science.

Data analysis is the science of analyzing/inspecting/cleansing/transforming/modeling raw data discovering useful information, informing conclusions, and supporting decision-making.

  • The techniques and processes of data analytics have been automated into mechanical processes and algorithms that work over raw data for human consumption.

Dara Analytics is a broader term embracing diverse types of data analysis. It is generally defined as “the systematic computational analysis of data or statistics”. It is used for the discovery, interpretation, and communication of meaningful patterns in data, applying data patterns to decision-making.

Data Analytics areas include machine learning, predictive analytics, prescriptive analytics, enterprise decision management, descriptive analytics, cognitive analytics, Big Data Analytics, retail analytics, supply chain analytics, store assortment and stock-keeping unit optimization, marketing optimization and marketing mix modeling, web analytics, call analytics, speech analytics, sales force sizing and optimization, price and promotion modeling, predictive science, graph analytics, credit risk analysis, and fraud analytics.

General types of Data Analytics

  1. Descriptive analytics: This describes what has happened over a given period of time.
  2. Diagnostic analytics: This focuses more on why something happened.
  3. Predictive analytics: This moves to what is likely going to happen in the near term.
  4. Prescriptive analytics: This suggests a course of action.

It is long used in marketing, credit risk assessment, fraud detection, manufacturing, healthcare, and government operations including law enforcement.

Data predictive analytics wiki article gives a long listing where it is used:

actuarial science, marketing, business management, sports/fantasy sports, insurance, policing, telecommunications, retail, travel, mobility, healthcare, child protection, pharmaceuticals, capacity planning, social networking and other fields.

Data science goes as an interdisciplinary field that uses scientific methods, processes, algorithms and systems to extract knowledge and insights from data, structured and unstructured. It is related to data mining, machine learning and big data, using techniques and theories from many mathematics, statistics, computer science, information science, and domain knowledge.

Data science is "to unify statistics, data analysis, informatics, and their related methods" to "understand and analyze actual phenomena" with data. It is pushed as a "fourth paradigm" of science, additionally to theoretical empirical and computational.

Where is the whole trick?

The trick is data analytics has been adopted by many sectors. Big Data analytics can require extensive computation, the algorithms and software used for BD analytics harness the most current methods in computer science, statistics, and mathematics. Just dubbed it under the most sexy name of AI, and sell it with any profit.

The travel and hospitality industry collect customer data and figure out where the issues lie and how to fix them.

Healthcare combines the use of high volumes of structured and unstructured data and uses data analytics to make quick decisions. The retail industry uses copious amounts of data to meet the ever-changing demands of shoppers to identify trends, recommend products, and increase profits.

The big tech companies and social networks services, as Facebook, and content companies use the same data analytics techniques to keep you clicking, watching, or re-organizing content to get another view or another click.

Gaming companies use data analytics to set reward schedules for players that keep the majority of players active in the game, etc.

So, when you are told that you enjoy some AI/ML/DL products or services keep in mind the following. It is a false advertisement of fake AI. Such an untrue or misleading information given to you to get you to buy something, or to come visit their site, platform or store. Those who make and sell fake ML/AI products must honestly present their products, services and prices to you.

For example, the IBM team sells its Summit supercomputer as a new generation of smart AI supercomputers, which has ZERO intelligence, no awareness of what it is doing and computing.

“Summit, an IBM AC922 system, links more than 27,000 NVIDIA Volta GPUs with more than 9,000 IBM Power9 CPUs to provide unprecedented opportunities for the integration of AI and scientific discovery.

Applying AI techniques like deep learning to automate, accelerate, and drive understanding at supercomputer scales will help scientists achieve breakthroughs in human health, energy, and engineering and answer fundamental questions about the universe.

The arrival of AI supercomputing and Summit means science has never been smarter”.

https://www.olcf.ornl.gov/summit/

Legally, you can’t blame the AI underworld in selling the fake ML/DL/AI product as far as the real-world AI with its definition, standards, specifications and technical parameters is formally adopted.

Sources

https://futurium.ec.europa.eu/en/user/10596

Critical Commentary on the “High-Level Conference on AI: From ambition to Action” Program

AI is in the critical stage of its revolutionary paradigm shift, as Transdisciplinary Science and Technology and Real-World Man-Machine Intelligence,…

Man-Machine Superintelligence Paradigm Shift: Trans-AI, Deep AI, or Real-World AI, as an integral Trustworthy AI Model

AI is becoming omnipresent and omniscient with virtually unlimited applications in all parts of everyday life. As such, it has many advantages,…

Trans-Technologies: AI as Transdisciplinary Science and Technology: how to solve the biggest challenges in the AI Industry

AI is set to change the ways the future world to work. What kind of future AI technology is to domineer will decide the kind of future world to live…

The Trans-AI will eat the world

Apple, Google, Amazon, Microsoft, and Facebook, enjoying the highest ever $1+ trillion market cap, are still working on an old and…