MIT Technology Review released the world’s top ten breakthrough technologies: the fastest vaccine technology is listed

By yqqlm yqqlm

Interview: Alibaba Cloud’s “Warm Cloud” theme event-30 million subsidies to help small and medium-sized enterprises break through the cold winter

In addition, this year’s list also includes green hydrogen energy, multi-skilled AI, remote technology, and lithium Metal batteries, digital contact tracking, ultra-high-precision positioning and data trust.

Liu Jiong, the Chinese co-publisher of MIT Science and Technology Review, said at the launch site of the list, “Whether it is a technology talent or a technology company, the core of it cannot be separated from breakthrough innovative technology. ”

He recalled that since 2001, “MIT Technology Review” has selected the “Top Ten Breakthrough Technologies” of that year every year. In 2009, Siri, which had not yet been acquired by Apple at the time, was selected; in 2013, “deep learning” was included in the list, and the outbreak period of 3-5 years was accurately given. In addition, the list also accurately predicted the rise of popular technologies such as brain-computer interfaces, smart watches, and cancer gene therapy.

“Although some of the top ten breakthrough technologies selected each year have not yet been widely used in large numbers at the time, and are even in the early stages of development, after time’s tempering and verification, many have been selected. Breakthrough technologies are standing on the cusp of commercial applications.” Liu Jiong said.

Massachusetts Institute of Technology Review was founded in 1899. It has long focused on the development and commercialization of emerging technologies. Since 2001, it has selected the “Top Ten Breakthrough Technologies” of the year. This year It is the 20th anniversary of the birth of this list.

MIT Technology Review released the world’s top ten breakthrough technologies: the fastest vaccine technology is listed

The following is the detailed interpretation of the list:< /p>

Significant significance: The effectiveness of the mRNA new crown vaccine is about 95%, and it has never been put into clinical application before, which may bring about a huge change in the field of medicine.

Globally, the new crown virus has claimed the lives of more than 2 million people, and several childhood friends of mRNA technology researcher Drew Weissman have died as a result.

Nowadays, several different types of new crown vaccines have been on the market. The new crown vaccine injected in the United States is completely dependent on Moderna’s vaccine and the vaccine jointly developed by BioNTech and Pfizer. The above two vaccines have one thing in common: both use mRNA technology.

The effective mechanism of mRNA vaccines is completely different from that of traditional vaccines. Traditional vaccines use live viruses, dead viruses, or parts of the virus shell to train the human immune system. The mRNA vaccine contains genetic material and is wrapped in liposomes. After being injected into the body, muscle cells absorb mRNA and produce a certain viral protein. The immune system will promptly produce antibodies and T cells to resist virus invasion.

Before that, although the research on mRNA has been for 20 years, it was not until last year that mRNA was used for the first time in drugs entering the market.

Significant significance: Learning large-scale computer models of natural language is a big step towards building AI that understands humans and interacts with the human world.

GPT-3 is a “large-scale language model” created by OpenAI, a research laboratory in San Francisco. The so-called “large language model” refers to an algorithm that uses deep learning to train words and phrases together through thousands of books and a large amount of text on the Internet. GPT-3 was officially released in 2020. It can imitate human-written text with an amazing degree of realism.

In the eyes of many people, this capability of GPT-3 can be regarded as a milestone on the road to true machine intelligence.

Significant significance: Faced with the abuse of personal data, data trust can help better manage data.

In the data age, while people’s lives are more convenient, privacy and security issues have become increasingly prominent. So, is there an organization that can protect people’s data rights in the same way as a union protects labor rights? And, like a doctor, can help people make wise decisions based on personal data management?

Perhaps, data trust is a feasible solution.

In the law, a trust refers to the act of trusting the trustee, the trustor proceeding from its own interests, and handing over the assets to the trustee for management. In a data trust, the trustee will manage the data or data rights of a group of people. Just as doctors have the responsibility to act in the interests of patients, data trustees manage the data or data rights of the principal and are also responsible for their interests. In theory, data trusts allow users to exercise their rights as data producers.

Data trust is not the only solution to issues such as privacy and security caused by data, but data governance that can control data and allow data sharing to benefit mankind is very necessary.

Great significance: Lithium metal batteries have high energy density, fast charging speed, and safe and reliable, making electric cars as convenient and cheap as gasoline cars.

Main researchers: QuantumScape Company, Carnegie Mellon University, Oak Ridge National Laboratory

One of the major problems restricting the development of the electric vehicle industry is battery technology.

Currently, lithium-ion batteries are commonly used in electric vehicles. These batteries are expensive, bulky, and have low energy density, and the liquid electrolyte they rely on can easily catch fire in the event of a collision. A series of shortcomings of batteries are embodied in electric vehicles: high price, low battery life, slow charging, and there are hidden safety hazards. These are the reasons why many car owners are discouraged from electric vehicles.

Obviously, to make electric cars more competitive than gasoline cars, a breakthrough battery is needed to make up for these shortcomings.

Silicon Valley startup QuantumScape claims to have developed a new lithium metal battery that uses solid electrolyte (ceramic) to overcome these shortcomings of traditional lithium-ion batteries.

Significant significance: Without obtaining personal location information, mobile phone users can learn whether they are in contact with people infected with the new coronavirus.

Main researchers: Apple, Google, etc.

The new crown epidemic that the world is experiencing in 2020 has caused “digital contact tracking” to attract people’s attention.

Under the new crown epidemic, technology has provided public health investigators with a new way of tracing the whereabouts of infected persons-digital contact tracing. Using this technology, health investigators no longer need to rely on the memory of patients to track their whereabouts, which reduces the pressure on disease monitoring.

This technology corresponds to the actual application and is called “Exposure Notification”. For the digital contact tracking system, programmers completed the establishment and operation within a few weeks, and shared the code as open source to ensure that users from all over the world, Apple and Android can use this function.

Specifically, the “exposure notification” does not track the user’s location, but uses Bluetooth to anonymously connect nearby mobile devices running the same application to ensure the anonymity and privacy of health data. As of January 2021, “MIT Technology Review” has learned about 77 “exposure notification” apps used by governments around the world.

Great significance: When the positioning technology is accurate to millimeter level or higher, it will create a new industry.

Main researchers: Institute of Aerospace Information Innovation, Chinese Academy of Sciences, ColdQuanta

The accuracy of the global satellite positioning system is improving from “meter” to “centimeter” level, which will be automatic Driving cars, delivery robots, etc., provide greater support for safe driving on the street.

The Beidou-3 global satellite navigation system officially opened in 2020 can capture the position changes of several meters on the ground in real time, and its processing accuracy can even reach the millimeter level. The system has been used to detect subtle changes in the surface of landslide-prone areas across China, and it was predicted that China’s Hunan Province would encounter the worst landslide in decades, allowing villagers to evacuate early. Experts from the Institute of Aerospace Information of the Chinese Academy of Sciences said that if the satellite positioning accuracy is still at the meter or decimeter level, this is impossible to achieve.

In fact, the further improvement of Beidou and Global Positioning System (GPS) accuracy requires ground facilities to improve positioning accuracy. Among the currently widely used methods, one is Real Time Kinematic (RTK) positioning with an accuracy of less than 3cm; the other is Precise Point Positioning (PPP), which can also achieve centimeter-level accuracy. . In addition, an expert from the Institute of Aerospace Information of the Chinese Academy of Sciences said: “We are developing PPP-RTK technology, combining the advantages of the two, and we are expected to put it into use in a few years.”

Great significance: During the 2020 epidemic Changes in the two important services of healthcare and education have had a huge impact on people’s overall well-being and quality of life. But the most important change is not the technology itself but our behavior, because teleconferencing and telemedicine have already existed.

Main researchers: Hong Kong online tutoring company Snapask, Homework Help, India Byju’s

In April 2020, the global epidemic reached its peak. Schools in more than 170 countries were closed, and The number of affected students is as high as 1.6 billion. Traditional schools in most regions of the world have switched to online teaching models, and Asia is no exception. For example, the demand for users of Snapask, an online tutoring company in Hong Kong, China, has soared.

Currently, Snapask has more than 3.5 million users in 9 Asian countries, which is double the number before the epidemic. Timothy Yu, who founded Snapask in 2015, said, “The number of users that we needed five years to reach before was realized in only one year due to the epidemic.”

Significant significance: “Multi-modality “The system can solve more complex problems, allowing robots to communicate and collaborate with humans in a true sense.

Main researchers: Allen Institute of Artificial Intelligence, University of North Carolina, OpenAI

At the end of 2012, artificial intelligence scientists figured out for the first time how to make neural networks “vision”, and then , They also mastered how to make neural networks imitate human reasoning, hearing, language and writing. Although artificial intelligence has become very similar to humans in accomplishing specific tasks, or even surpass humans, it still does not have the “flexibility” of the human brain, that is, the human brain can learn skills in one situation and apply them to another. In a situation.

For example, although AlphaGo can beat the best Go players in the world, it cannot extend this strategy beyond the board, and cannot understand and adapt to the ever-changing external world.

In addition, the design of the artificial intelligence system is to do only one thing at a time. For example, computer vision and audio recognition algorithms can perceive things, but they cannot use language to describe them; natural language models can manipulate text, but text is out of any sensory reality.

In this regard, the growth process of children is a good example. Children understand the world by “perceiving” and “talking about” the world. When they begin to associate words with scenery, sounds, and other sensory information , They can describe more and more complex phenomena and dynamics, and build a complex world model, which can help them navigate unfamiliar environments and incorporate new knowledge and experience into them.

Inspired by the growth process of children, if the senses and language are combined, and artificial intelligence has a way closer to humans to collect and process information, can it develop an understanding of the world? The answer is yes. These “multimodal” systems that can simultaneously obtain human intelligence’s senses and language should generate a more powerful artificial intelligence, and it will be easier to adapt to new situations and solve new problems.

Significant significance: TikTok can not only accurately recommend videos of interest to users, but also help them expand into new areas that have intersections through recommendation algorithms.

Main researcher: TikTok

TikTok is one of the most attractive and fastest-growing social media platforms in the world. As of now, TikTok has more than 2.6 billion downloads worldwide and has 100 million users in the United States. TikTok’s unique way of discovering and delivering content is its attractive “secret weapon”.

TikTok mixes the videos of internet celebrity bloggers and newcomer bloggers on the “Recommended for you” page, and then rewards high-quality creative content with page views. In this way, more new bloggers will be attracted The video is pushed to the majority of users.

In a nutshell, TikTok’s skills to enhance user stickiness are becoming more and more sophisticated. It can not only accurately recommend videos of interest to users, but also help them expand into new areas that have intersections through recommendation algorithms. .

Significant significance: Green hydrogen is a green, carbon-neutral energy source, an expansion of renewable wind and solar energy, and may become the core fuel for low-carbon in the future.

Key researchers: Green Hydrogen Alliance ThyssenKrupp Group, International Energy Agency, McKinsey Consulting Company

Maturity: Expected in 2030

Most Hydrogen is produced by combining natural gas with high-temperature steam, which is the so-called gray hydrogen. Green hydrogen is produced through water electrolysis, and the current production cost is about three times that of gray hydrogen.

With the continuous decline in the cost of wind and solar energy, and the emergence of economies of scale in green hydrogen production, the cost of hydrogen production is also expected to usher in a substantial reduction. With the development of carbon capture technology, more hydrogen can be extracted from natural gas without emitting a large amount of carbon dioxide, and green hydrogen may also become the core fuel for low carbon in the future.

On the one hand, the fuel cell method uses hydrogen and the reverse reaction of water electrolysis is used to generate water and electricity through the reaction of hydrogen and oxygen without producing nitrogen oxides. On the other hand, by combining hydrogen and carbon, it is possible to produce liquid synthetic hydrocarbon fuels that are easier to handle than hydrogen. This type of liquid fuel is cleaner and can replace gasoline or diesel.

Hydrogen can also be used to store electricity produced by renewable energy sources. When the wind weakens, clouds cover the sun, or electricity demand increases, this energy can be converted back into electricity and fed into the grid.

The International Energy Agency (International Energy Agency) predicts that by 2050, hydrogen can provide more than 10% of global energy demand and produce more than 11 million gigawatt-hours of energy each year.