-
26 August 2019 - updated 1 year ago -
0 comments
Global AI research performance could be analyzed from a total number of scientific papers published and patent applications as filed.
It is rather naive to think that “AI is being shaped by nine companies, all in the US or China: Alibaba, Amazon, Apple, Baidu, Facebook, Google, IBM, Microsoft, and Tencent", which could be clustered as G-MAFIA and BAT-Triada.
It is like presented by Forbes' hired writers (The Big Nine: How the Tech Titans and Their Thinking Machines Could Warp Humanity; Nine Companies Are Shaping The Future Of Artificial Intelligence, etc.)
This massive cognitive bias formed by the G-MAFIA and BAT-Triada propaganda machine, their misinformation, disinformation and “fake news”, as Trump likes to say.
Key trends in AI research of narrow/weak automated AI/ML/DL could be well seen from WIPO Technology Trends 2019 Artificial Intelligence. https://www.wipo.int/publications/en/details.jsp?id=4386
As the evolution of AI patent applications and scientific publications shows, nearly 340,000 patent families and more than 1.6 million scientific papers were published from 1960 until early 2018.
Universities contribute significantly to AI research in specific fields, with Chinese universities dominating.
There are 167 universities and public research organizations ranked among the top 500 patent applicants. Of these, 110 are Chinese, 20 are from the U.S., 19 from the Republic of Korea and 4 from Japan. Four European public research organizations feature.
The U.S. and China are the two most popular offices for filing AI patents, in line with patenting trends in other fields, followed by Japan. These three offices account for 78 percent of total patent filings.
Chinese organizations make up 17 of the top 20 academic players in AI patenting as well as 10 of the top 20 in AI-related scientific publications. Chinese organizations are particularly strong in the emerging technique of deep learning. The leading public research organization applicant is the Chinese Academy of Sciences (CAS), with over 2,500 patent families and over 20,000 scientific papers published on AI. Moreover, CAS has the largest deep learning portfolio (235 patent families). Chinese organizations are consolidating their lead, with patent filings having grown on average by more than 20 percent per year from 2013 to 2016, matching or beating the growth rates.
The number of patent applications filed annually in the AI field grew by a factor of 6.5 between 2011 and 2017.
The AI techniques on which the patent literature focuses most extensively are machine learning, followed by logic programming (expert systems) and fuzzy logic. The most predominant AI functional applications are computer vision, natural language processing and speech processing.
Machine learning is the dominant AI technique disclosed in patents and is included in more than one-third of all identified inventions.
IBM has a portfolio of AI patent applications with 8,290 inventions, followed by Microsoft with 5,930, Toshiba (5,223), Samsung (5,102) and NEC (4,406).
In certain techniques and fields, the highest numbers of patent applications originate from companies with a high degree of specialization and expertise in that field. Examples include Baidu, which ranks highly for deep learning, Toyota and Bosch, which are prominent in transportation, and Siemens, Philips and Samsung in life and medical sciences.
Acquisitions make the big-tech bubbles AI research and IP strategies.
In total, 434 companies in the AI sector have been acquired since 1998, with 53 percent of acquisitions having taken place since 2016. The number of acquisitions identified in the AI sector has increased every year since 2012, reaching 103 in 2017.
Although Alphabet (including Google, DeepMind, Waymo and X Development) ranks 10th in the number of inventions filed, with 3,814 in total, it ranks 1st in terms of acquisitions of AI companies.
Apple and Microsoft have also been active in acquisitions.
Certain companies, such as IBM and Intel, target mature companies. The majority of acquired companies are, however, startups with small or non-existent patent portfolios.
Categorization of AI technologies:
AI techniques: advanced forms of statistical and mathematical models, such as machine learning, fuzzy logic and expert systems, allowing the computation of tasks typically performed by humans; different AI techniques may be used as a means to implement different AI functions.
AI functional applications: functions such as speech or computer vision which can be realized using one or more AI techniques.
AI application fields: different fields, areas or disciplines where AI techniques or functional applications may find application, such as transportation, agriculture or life and medical sciences.
Looking first at trends in AI techniques, machine learning predominates, representing a massive 89 percent of filings mentioning this AI technique and 40 percent of all AI-related patents.
Trends in AI techniques
Looking first at trends in AI techniques, machine learning predominates, representing a massive 89 percent of filings mentioning this AI technique and 40 percent of all AI-related patents. Machine learning grew by 28 percent from 2013 to 2016; in the same period, fuzzy logic has grown by 16 percent and logic programming by 19 percent.
Within machine learning, every AI technique showed an increase in annual filing numbers for the same period, but some stand out. Deep learning is the fastest growing technique in AI, with an 175 percent increase over the period. Multi-task learning, the next fastest, grew by 49 percent. Other techniques with notable increases were neural networks, latent representation and unsupervised learning.
Trends in AI functional applications
Turning to trends in AI functional applications, computer vision, which includes image recognition, is the most popular. Computer vision was mentioned in 49 percent of all AI-related patents and grew by 24 percent during 2013 to 2016. The other two top areas in functional applications are natural language processing (14 percent of all AI-related patents) and speech processing (13 percent).
While these three functional applications are the most important in terms of the total number of filings, others are emerging and growing fast. AI filings concerning both robotics and control methods have increased by 55 percent, for example, while those for planning/scheduling have grown by 37 percent.
Within computer vision – the top functional application – biometrics has seen an average annual growth rate of 31 percent and scene understanding one of 28 percent. Within natural language processing, semantics has grown by 33 percent and sentiment analysis by 28 (though it still only accounts for 1 percent of natural language processing applications). Within speech processing, speech-to-speech has grown by 15 percent, and speech recognition and speaker recognition have both grown by 12 percent.
Trends in AI application fields
Lastly, in AI application fields, the top industries are transportation (15 percent of all AI-related patents), telecommunications (15 percent), and life and medical sciences (12 percent). Growing industries are transportation, agriculture, and computing in government, with annual growth rates of at least 30 percent between 2013 and 2016.
Looking at trends over ten years, the boom in transportation technologies becomes more evident: representing just 20 percent of applications in 2006, by 2016 it accounted for one-third of applications (with more than 8,700 filings).
Telecommunications, the second most important application field, has remained at around 24 percent during this period.
WIPO Technology Trends 2019 – Artificial Intelligence
Up to now, all the techniques and applications refer to individual tasks performed by AI systems, known as “narrow AI.”
This is to be distinguished from general AI.
It is Strong AI, True Intelligence, Real AI or Artificial General Intelligence (AGI), a form of machine intelligence that is equal or much superior to human intelligence.
Key characteristics of strong AI include the ability to perceive and conceive, reason, solve problems, make judgments, plan, teach, learn and communicate and interact with the world. It should also have deep understanding, consciousness, thoughts, self-awareness, sentience, and sapience.
Real AI could be grown a human mind, starting with a childlike mind and developing an adult mind through teaching and learning, while interacting with the world and learning from it, acquiring its own knowledge and experience, common sense and language.
The multi-trillion question is when we are to develop True AI, Real AI, or General AI (gAI).
The gAI paradigm shift, a fundamental change in the basic concepts and experimental practices of AI Science and Technology, is to take place within next 5 years.
Paradigm shifts arise when the dominant paradigm under which current R&D science operates is rendered incompatible with new demand, facilitating the adoption of a new model/theory/paradigm of General AI.
What is certain:
gAI will penetrate everywhere and everything, banking, retail, healthcare, manufacturing, telecommunications, transportation and automotive, logistics, warehousing/transportation/delivery, physical and social infrastructure, education, job, government, etc.
Real AI Manifesto: Artificial Global Intelligence (AGI)
"Whoever Creates Real Artificial Intelligence Will Rule the World"
https://www.linkedin.com/pulse/g...
Universal computing ontology as applied to human minds and general AI:
https://www.igi-global.com/book/...
Kiryl Persianov's answer to What are the major types of artificial intelligence?
Kiryl Persianov's answer to What are some artificial intelligence breakthroughs made by individuals?