Source: "Global" magazine
Release time: 2017-04-24
With the development of optomechanical integration, bioengineering, biochemical robots, and the comprehensive progress of system science, the autonomous logical thinking of computers will have sufficient behavioral performance, and then truly break away from the complete control of humans and have autonomous intelligent thinking.
It is almost foreseeable that in the next 5 to 10 years, or even longer, big data, brain-computer interfaces and neural engineering will become key targets for scientists in artificial intelligence and related fields to make breakthroughs. Perhaps in the not-too-distant future, the human-machine-cooperative biochemical man can really fly freely like Iron Man, freely manipulating "weapons" like a mecha warrior, rewriting the trajectory of life, traveling through time and space, and even exist without material attributes.
In an era when the world is covered by the Internet, and biochemical, bionic, and human coexistence, a semi-robot Motoko Kusanagi with a human brain and soul, designed and created by Dr. Olette, joined the elite force "Nine Public Security Lessons" Assisting the government in fighting high-tech crimes... The Hollywood sci-fi blockbuster "Ghost in the Shell", which is currently being screened, not only tells the story of cyborgs searching for their souls, but also illustrates the relationship between humans and machines in the future.
Human-machine integration is regarded as the next important turning point of artificial intelligence, and American technology madman Elon Musk is working to promote human-machine integration. On March 28th, Musk announced the establishment of a company called "Neural Connection". This technology company is committed to connecting the human brain with computers and improving human brain capabilities to ensure that humans have the ability to deal with artificial intelligence (AI) The threat posed.
The former science fiction is approaching reality, with the complex mentality of expecting and worrying about the future of artificial intelligence, people can't help asking: how far is human beings from realizing the integration of man and machine?
The threat of artificial intelligence
With the development of optomechanical integration, bioengineering, biochemical robots, and the comprehensive progress of system science, the autonomous logical thinking of computers will have sufficient behavioral performance, and then truly break away from the complete control of humans and have autonomous intelligent thinking.
Due to the development of artificial intelligence and the progress of software and hardware, the logical analysis ability of computers has been greatly improved, until the comprehensive logical analysis of computers is improved to logical thinking. This logical thinking can independently generate new logic according to environmental conditions and get rid of the human framework. Control becomes a kind of autonomous intelligent thinking. Scientists even put forward a prediction: before the end of the 21st century, humans will no longer be the most intelligent species on earth.
It is based on concerns that artificial intelligence surpasses humans, Musk officially launched the "neural connection" plan. The Wall Street Journal reported on March 27 that the "Neural Connection" company will use a technology called "Neural Net" to implant tiny electrodes in the human brain to establish connections with computers. This technology can not only treat brain dysfunction diseases such as epilepsy and Parkinson's disease, but also improve the human brain's ability: "download" human thinking to the computer, or "upload" information from the computer to the human brain through electrodes. Human intelligence and artificial intelligence are effectively integrated to improve human cognitive ability and memory.
In a Twitter article on March 28, Musk expressed his concern that artificial intelligence may threaten humans, thinking that "this kind of risk is so high that we have to conduct related research." Musk warned at a conference in California last year that if humans create "super-intelligent artificial intelligence products" that far surpass humans in all aspects, then humans may become "house cats" in the face of powerful artificial intelligence. . Musk proposed the "neural network" at the time, saying that this technology can improve the capabilities of the human brain.
Professor Nick Postrom of the University of Oxford in the United Kingdom proposed in his book "Superintelligence" that artificial intelligence will change in the direction of superintelligence. Many people in academia and industry have expressed similar concerns as Musk. A voice represented by British scientist Stephen Hawking and Microsoft founder Bill Gates is: "If a machine develops enough intelligence in a few years, it will eventually become a major concern for mankind."
Hong Xiaowen, senior vice president of Microsoft and chairman of Microsoft Asia Pacific R&D Group, believes that AI is a tool and technology created by human beings. No one will have no AI because AI's ideas and algorithms all come from humans. The core competitiveness of mankind is innovation and creation. Throughout the history of creation, the consistent strategy adopted by mankind is to make bold assumptions and carefully verify. Among them, the creativity of bold assumptions comes from humans, and the task of careful verification is given to the computer. In the process of verification, we repeatedly modify our assumptions and modify our ideas, and finally we can create new things. "If we say that there must be superintelligence in society in the future, it must be achieved through the integration of human intelligence (HI) + artificial intelligence (AI)."
After a long period of evolution, humans have long learned from historical experience: the fewer enemies the better. If the AI shows divine power, you might as well join hands and turn it into a friend. This kind of future where human intelligence and artificial intelligence work together is probably the present of the integration of personal computers that Musk and other tech madmen are laying out.
Three steps of human-machine integration
Pan Yunhe, former executive vice president of the Chinese Academy of Engineering, pointed out in an interview with a reporter from "Global" magazine that it is important to use computers to simulate human intelligence, but letting computers work with humans to complement each other and become a kind of "1+1>2" enhancement Sexual intelligence systems are even more important. At present, various wearable devices, intelligent driving, exoskeleton devices, and human-machine collaborative surgery have emerged one after another, and there is more room for human-machine collaboration in the macro system, which indicates a bright future for human-machine collaboration to enhance intelligent systems.
Smart devices are embedded in the body to read physiological data in real time. Machines understand people better than humans... This judgment comes from the recent hot-selling book "A Brief History of Mankind". The author Yuval Harari believes that with artificial intelligence and biological With the rapid development of technology, the collaborative integration of man and machine will be fully realized in this century, and human life will undergo tremendous changes in the future.
As early as 15 years ago, the cybernetics expert and professor at the University of Reading in the United Kingdom, Kevin Woik took the lead in proposing and analyzing the concept of Cyborg in his book "Biorobots and Me". The literal meaning of the corresponding phrase in English is the combination of organic life and inanimate machinery.
It is worth mentioning that in order to write academic works, this professor used surgery to implant a silicon chip pulse transponder into his left arm in 1998, and in 2002 he implanted a square chip with a side length of 3 mm. He entered his left wrist and connected his nervous system to the computer through a chip line. He also said in the book that when the man wearing the chip was in and out of the office building, all the rooms controlled by the computer could recognize him. , Automatically turn on the lights and open the door, the computer will also send out "Good morning, Professor Warwick" greetings, and notify him when receiving a new mail, and automatically download the new mail, his assistant can also find him at any time.
In fact, since the birth of artificial intelligence, it has tried to improve and extend human capabilities in all aspects. The process of human-computer integration has begun and is still accelerating. As a deep cultivator in the field of artificial intelligence, Dai Wenyuan, the founder and CEO of the Fourth Paradigm, believes that the collaborative integration of humans and machines can be divided into three development stages, namely the perception fusion stage, the behavior fusion stage and the thought fusion stage.
The first stage: perception fusion.
In fact, we have passed a more mature stage of perception fusion. In this stage, sensors appear as core components and are used in different fields and technologies, such as the familiar smart home, visual recognition technology, and voice recognition technology. With the help of artificial intelligence, human perception has been magnified and expanded hundreds to thousands of times, and has been given the ability to "see six directions, hear all directions". At the same time, artificial intelligence has completed the accumulation of technology at this stage, laying the technical premise for the next development.
The second stage: behavior integration.
At present, the most concerned technology in the field of artificial intelligence is the technology that can be implemented in the industry. And this is the stage of behavior fusion, the essence of human-computer fusion-based on the learning and imitation of human "teacher", the machine tirelessly trains and updates, and finally realizes the supplement and complement of humans in certain behavioral or work capabilities. Beyond. For example, in the financial field, artificial intelligence can effectively alleviate the shortage of manpower and energy in multiple scenarios such as precision marketing, risk prevention and control, and robo-advisory. Take robo-advisors as an example. In the past, financial institutions could only provide a small number of VIP investors with personalized financial management services due to high labor costs. Now, the machine can provide customers with more thoughtful and meticulous financial management advice by learning the investment advice criteria and experience of the customer manager, and then through its own ultra-high-dimensional model, and win the trust of customers. Many industry experts believe that in the stage of behavior integration, the R&D gap between China and the United States and other developed countries is constantly narrowing. It can be said that this stage is the best time for China to achieve corner overtaking.
The third stage: integration of ideas.
The explanation of the stage of fusion of thoughts is actually quite similar to the viewpoint of "A Brief History of Mankind" cited above. Regarding this stage of exploration, whether it is domestic or international, it is still in its infancy, and there is not much mature theory and practice. Perhaps in the future, with the development of machine learning, biosensors, brain-computer interaction and other technologies, this fusion process will accelerate. The artificial intelligence program can fully study you every day after your birth, from every short message to every second of your heartbeat, and the artificial intelligence that has won your heart will eventually merge with your thoughts and make more beneficial choices for you It can range from shopping as a brand choice to a tangled lifelong event like marriage.
Human-machine-object ternary integration
Science fiction writers and science fiction fans also bravely imagined: When humans no longer focus on the evolution of their own limbs, but connect their brains and limbs with computers and machines, when knowledge and external information are input into the human brain through neuron signals at any time, And save it for recall. Humans can have a genius computer without using their brains to learn.
In theory, what humans think can also be immediately parsed by a computer in the form of signal input and converted into cross-media symbols, which are displayed in the form of videos or images; connect the signals of different humans without language or gesture communication You can have the ability to "read mind" and realize "transparency of thought."
City managers propose a development path in accordance with the speed and direction of technological progress: intelligent demand traction, enabling AI to develop in cities, medical care, transportation, logistics, robots, unmanned driving, mobile phones, games, manufacturing, society, economy, etc. China has become a new technology and new target, and many companies and cities have already deployed AI.
The pursuit of "simulating humans with computers" in the past has been transformed into an enhanced hybrid intelligence that combines machines and people; a new group intelligence system is formed by the combination of machines, people, and networks; and a combination of people, machines, networks, and things Smart cities and other more complex smart systems.
Pan Yunhe believes that the world has evolved from a binary spatial structure formed by humans and objects to a three-dimensional spatial structure composed of people, objects, and information grids. The interaction between people, things, and the network will form a variety of new calculations, including perception fusion, "people in the loop" (the operator will still have the opportunity to perform a second or even uninterrupted command after the first command input Corrections), augmented reality, and cross-media computing.
When the fire of AI2.0 ignites, how to continue the firewood for the new era? The scientists interviewed by reporters from "Global" magazine finally formed a more consistent answer: The first aid is data of sufficient scale and high quality.
Although the computing speed of computers has been increasing exponentially in the past half century, there are more and more things that can be done, but people feel that they are fast but not smart enough, such as being unable to answer people’s questions, playing chess, and not knowing. People cannot drive and are not good at actively making judgments... Associate Professor Qin Zengchang of Beijing University of Aeronautics and Astronautics believes that when the amount of data is large enough, many intelligence problems can be transformed into data processing problems. Because from the methodological point of view, the way that computers acquire intelligence is different from human intelligence that is good at logical analysis and reasoning. Artificial intelligence makes more use of big data and learns to obtain information and knowledge from data.
Yu Kai, co-founder and chief scientist of Suzhou Spitz, proposed that the opportunities for the development of artificial intelligence are the increase in data volume; the second is the emergence of cloud computing. "The ability of computers to obtain such high intelligence today depends largely on data." Yu Kai said.
A more far-reaching boost is the research breakthrough in cognition and brain science. The book "New Trends in Technological Development and Strategic Choices for 2020" points out that fundamental breakthroughs in information technology may depend on the progress of brain science. Once a major breakthrough is made in brain science, it will surely cause a revolutionary change in information technology.
Brain science and neurotechnology are one of the fastest-growing disciplines in my country in the past 20 years, and they are also key areas where the United States and Europe have successively made strategic arrangements. The team of Jiang Tianzi from the Brain Network Group Research Center of the Institute of Automation of the Chinese Academy of Sciences has successfully drawn a new human brain map, which includes 246 sub-regions of fine brain regions and multi-modal connection patterns in the sub-regions of the brain. The director of the research center and doctoral supervisor Jiang Tianzi introduced that the brain network map is an inevitable trend in the development of the human brain map and the progress of neurotechnology. It is the support for the development of brain science, cognitive science, and cognitive psychology. It is bound to be artificial intelligence. A breakthrough in the field laid the foundation.
The cooperation between neuroscience and informatics will penetrate deeply into the field of artificial intelligence. Neural networks-models built by computers imitating human neurons will accelerate machines with many skills. It is almost foreseeable that in the next 5 to 10 years, or even longer, big data, brain-computer interfaces and neural engineering will become key targets for scientists in artificial intelligence and related fields to make breakthroughs.
Perhaps in the not-too-distant future, the human-machine-cooperative biochemical man can really fly freely like Iron Man, freely manipulating "weapons" like a mecha warrior, rewriting the trajectory of life, traveling through time and space, and even exist without material attributes.
1. The domestically-made alternatives are intangible? Come back to the Zhaoyi Innovation Live Class!
2. Can the open source RISC-V be the antidote to China's "core shortage"?
3. Raspberry Pi Pico: MCU for only $4
4. There are many reasons why MCU supports AI function~
5. In 2020, 20 software engineering principles I learned~
6. The application of state machine ideas in embedded development~
Disclaimer: This article is reproduced online, and the copyright belongs to the original author. If you are involved in copyright issues, please contact us, we will confirm the copyright based on the copyright certification materials you provide and pay the author's remuneration or delete the content.