Artificial Intelligence or Digital Superintelligence as a Global Scale Platform and General Purpose Technology
Artificial Intelligence or Digital Superintelligence as the next Global Scale Platform and the last General Purpose Technology
A goal of AI (Artificial Intelligence) or DI (Digital Superintelligence) is not to be equal or exceed human intelligence, but to become the last and the best of global scale platforms (GSPs) or general purpose technologies (GPTs).
GPTs are technologies that can affect an entire economy at a global level revolutionizing societies through their impact on pre-existing economic and social structures.
GPTs' examples: the steam engine, railroad, interchangeable parts and mass production, electricity, electronics, material handling, mechanization, nuclear energy control theory (automation), the automobile, the computer, the Internet, medicine, space industries, robotics, software automation and artificial intelligence.
The four most important GPTs of the last two centuries were the steam engine, electric power, information technology (IT), and general artificial intelligence (gAI).
And the time between invention and implementation has been shrinking, cutting in half with each GPT wave. The time between invention and widespread use for the steam engine was about 80 years; 40 years for electricity, and about 20 years for IT
Now the implementation lag for gAI-related technologies will be about 5-10 years.
As Vladimir Putin warned Russians that the country that led in technologies using general artificial intelligence will dominate the globe.
AI = General AI = Strong AI = Full AI = ASI = Digital Superintelligence = Superintelligence =True AI = Real AI
There is no other kind of AI, but strong and general, full and complete, or true and trustworthy, and all the rest is just simulation and imitation, or a fake AI.
Oftentimes, I preach that there is only one AI, Real and True AI vs. Fake and False AI, as today's Big Tech ML & DL, as far as a software automation with no intelligence at all could pass the Turing test.
The last has a broad range of ways in which it can be applied – from chatbots to predictive analytics, from recognition systems to autonomous vehicles, and many other special domains and use cases and narrow tasks, whatever liable to automation.
Rethinking Weak Vs. Strong AI | Snowdrop Solution
Weak or Narrow, Strong or General, or Supraintelligent AI, it is rather a fiction. It is like with human general intelligence/broad intellect, one either has it or not, as it is measured by universal psychometrics and various IQ tests.
Universal psychometrics: Measuring cognitive abilities in the machine kingdom
So, a big question is how to measure real AI, general intelligence that can handle any task or problem in any domain. Or, how to measure its efficiency, software efficiency or hardware efficiency; algorithmic efficiency or engineering productivity; data processing efficiency or general intelligence efficiency.
Efficiency is measured as the ratio of useful output to total input, r=P/C, where P is the amount of useful output ("product") produced per the amount C ("cost") of resources consumed.
AI efficiency shouldn't be confused with AI effectiveness. AI Effectiveness is the simpler concept of being able to achieve a desired result.
Much depend on how you view AI, as the machine intelligence software/hardware to interact with its environments, real or virtual, using
- predetermined rules and search algorithms,
- pattern recognizing machine learning models,
- neuro-symbolic models,
- causal data models, as the mental models of reality,
- all to make recommendations, predictions, or decisions based on those analyses.
There are generally three approaches to the AI performance measures:
- Simply naive;
- Human-like;
- Integrally Complex
The first method is exploited by OpenAI.
Our results suggest that for AI tasks with high levels of recent investment, algorithmic progress has yielded more gains than classical hardware efficiency.
Algorithmic improvement is a key factor driving the advance of AI. It’s important to search for measures that shed light on overall algorithmic progress, even though it’s harder than measuring such trends in compute.
44x less compute required to get to AlexNet performance 7 years later
The second one is about Intelligence as a collection of task-specific skills or Intelligence as a general learning ability.
A fundamental notion in psychometrics is that intelligence tests evaluate broad cognitive abilities as opposed to task-specific skills.
The hallmark of broad abilities (including general intelligence is the power to adapt to change, acquire skills, and solve previously unseen problems – not skill itself, which is merely the crystallized output of the process of intelligence. Testing for skill at a task that is known in advance to system developers (as is the current trend in general AI research) can be gamed without displaying intelligence, in two ways: 1) unlimited prior knowledge, 2) unlimited training data. To actually assess broad abilities, and thus make progress toward flexible AI and eventually general AI, it is imperative that we control for priors, experience, and generalization difficulty in our evaluation methods, in a rigorous and quantitative way.
The intelligence of a system is a measure of its skill-acquisition efficiency over a scope of tasks, with respect to priors, experience, and generalization difficulty.
https://arxiv.org/pdf/1911.01547.pdf
The third one is an integral approach to develop an aggregate performance measure viewing AI model as a whole intelligent entity.
Algorithmic efficiency, accessible computing equipment and software; the number of computational resources used by the algorithm computation time, the number of steps necessary to solve a problem, and memory space, the amount of storage needed while solving the problem, as in general hierarchies of complexity classes.
Data efficiency
General intelligence efficiency
Engineering productivity.
As a result, we don't have the true AI tech we need today, in the pandemic world.
The pandemic has been proving how fragile the society we've built is, its public-health systems, food supply chains, education infrastructure, cybersecurity, job markets, information networks, electoral machinery, and more.
The next decade will likely bring more crises...and we aren't equipped with a global-scale true AI technology, as strong AI models, algorithms, software applications, robots, machines, hyperscale platforms, and digital ecosystems.
Most of the companies with the world’s highest market capitalizations are tech companies that generate much of their revenue from the false AI-based digital ecosystems, as B2C plays or B2B spaces. For example, Amazon monetizes on combining e-commerce, cloud computing, logistics, and consumer electronics, while China’s Tencent provides services including social media, gaming, finance, and cloud computing, all with advanced analytics tools to exploit personal information to personalize products and services as never before.
It is estimated that at least a dozen sectors, including B2B services, mobility, travel and hospitality, health, and housing, are reinventing themselves as vast ecosystems, networks of networks that could add up to a $60 trillion integrated network economy by 2025.
For in the emerging world of Ecosystem 2.0, data and AI tools are the holy grail.
Ecosystem 2.0: Climbing to the next level
It must be clear for all of us that Real Artificial Intelligence is a leading force of change, which is to play a critical role in shaping the world to come, as more just, intelligent, equitable, and secure for all.
Thus, fast building the strong and general, full and complete, or true and trustworthy AI is a matter of life or death for humanity, and most global innovation investment should be focused on the next big innovation.
Resources
Why AI is to take over the human world
https://www.dhirubhai.net/pulse/why-ai-take-over-human-world-azamat-abdoullaev/?published=t
How to Regulate the Big Tech fake AI: "AI and ML Falsification Act"
https://www.dhirubhai.net/pulse/how-regulate-big-tech-fake-ai-ml-falsification-act-azamat-abdoullaev/?published=t
How to remedy Fake AIs and build Real AI World: True Trustworthy AI
https://www.dhirubhai.net/pulse/how-remedy-fake-ais-case-big-tech-azamat-abdoullaev/?published=t
What is the (true) nature of reality?
https://www.dhirubhai.net/pulse/what-true-nature-reality-azamat-abdoullaev/?published=t
GPT Sources
AI @ Facebook Overview
AI as a GPT ? Adoption to drive innovation across sectors ? Will generate major social benefits and improve welfare/productivity ? Spillovers throughout economy as w/ previous general purpose technologies
“Artificial intelligence, especially machine learning, is the most important general-purpose technology of our era. The impact of these innovations on business and the economy will be reflected not only in their direct contributions but also in their ability to enable and inspire complementary innovations. New products and processes are being made possible by better vision systems, speech recognition, intelligent problem solving, and many other capabilities that machine learning delivers.” Erik Brynjolfsson & Andrew McAfee (Harvard Business Review, 2017)
Conceptualizing AI as a General Purpose Technology
“For more than 250 years the fundamental drivers of economic growth have been technological innovations…… The most important general-purpose technology of our era is artificial intelligence, particularly machine learning.” —Erik Brynjolfsson and Andrew McAfee, 2018
AI for Good Breakthrough Days: Collective Pandemic Intelligence
https://aiforgood.itu.int/breakthrough-days/