The wave of intelligence does not retreat 2018 AI bears the brunt

If 2017 is the first year of artificial intelligence (AI) development, then in the new 2018, artificial intelligence will begin to go out of the laboratory and into more practical areas of technology and applications. With the artificial intelligence technology entering the chip products, it is expected that 2018 will be a year of artificial intelligence. . .

In 2017, the semiconductor industry has a good development, and the market sentiment is quite satisfactory for many manufacturers. 2018 The new year is coming, what will happen to the development of the entire semiconductor technology industry? What are the key technologies of the mainstream market? How should the industry layout the market for the next year? The following will come together.

From the perspective of the overall industry, Gartner lists the top ten technology trends in 2018, including artificial intelligence (AI) based, intelligent application and analysis technology, smart objects, digital avatars, cloud-to-edge computing, and conversational platforms. , immersive experience, blockchain, event-driven, and continuous adaptation to risk and trust.

Gong Peiyuan, senior partner of Gartner Greater China, said that the above ten technological trends may be similar to the 2017 forecast, but it also represents a more mature trend of each technology, which in turn affects industry trends and people. Life has also become a key technology in 2018.

The top ten technology trends proposed by Gartner are in fact related to the Intelligent Digital Mesh. Gong Peiyuan explained that the intelligent part is undoubtedly artificial intelligence. In 2017, the industry regarded it as the first year of artificial intelligence development. By 2018, Gartner believes that artificial intelligence will enter the field of practical technology and become a usable technical achievement. Instead of being the technology researched in the laboratory, 2018 will be a year of artificial intelligence.

Therefore, in 2018, artificial intelligence will enter the chip product, namely "AI on Chip". For example, the technology industry has begun to have greater interest in the use of graphics processor (GUP)-based servers; Google has developed its own dedicated chip Tensor Processing Units (TPU) for its own server center; Microsoft (Microsoft) released FPGA-based Low-latency deep learning cloud platform Project Brainwave; Intel's Nervana chip for machine learning; Nvidia's graphics processing unit (GPU) and artificial intelligence chips are constantly evolving; Qualcomm launches neural network The chip; IBM's first brain chip, TrueNorth, and the neuron adaptive processor Spiking, which was developed for unsupervised brain imaging, and the startups Cerebras and Groq, are expected to launch artificial intelligence chips in 2018.

Gong Peiyuan believes that artificial intelligence has begun to prove to the public or the technology industry that its future influence is quite far-reaching, and will attract IT or chip "big old" and startups to actively launch artificial intelligence-related chip products. For example, artificial intelligence will create new applications in the "user experience" in 2018, that is, a new user experience brought out by Apple iPhone X in 2017, and then through manual Intelligently enhances the user data analysis process, reduces blind spots, and increases the ability to analyze specific aspects, giving users a new and different experience, and artificial intelligence can also help service providers to better meet consumer needs and better. Application service.

Figure 1: Chip architecture with built-in natural user interface for future smart objects

Furthermore, the mainstream application of artificial intelligence will emerge in 2018, especially for applications related to “security”. As artificial intelligence needs to collect more and more data as the basis of its in-depth analysis, artificial intelligence not only needs to protect the security of data, but also has the ability to prevent and find errors or false information, so as to avoid errors in the analysis results.

Not only that, but another focus of the development of artificial intelligence in 2018 is the interaction of people with artificial intelligence. Gong Peiyuan explained that at this stage, people still have doubts about artificial intelligence, especially whether artificial intelligence will replace artificial. In fact, if people or enterprises really understand the "use" of artificial intelligence, they will not be replaced, but they can also be accepted. 1+1 is greater than 3.

That is to say, from now until 2019, the artificial intelligence will replace the number of jobs created by it; however, by 2020, Gartner predicts that the job creation created by artificial intelligence will be enough to make up for the number of replacements. What's more, artificial intelligence will increase the productivity of many jobs. If used ingeniously, it can really enrich people's employment life, re-engineer old jobs, and create new industries.

Among the top 10 ICT trends proposed by the Institute of Industrial Intelligence (MIC) in 2018, there are several related to artificial intelligence. Hong Chunhui, consultant and director of the MIC industry, said that the development of the ten key trends in 2018 is actually centered on four directions: intelligence, openness, service and integration. The development of the industry towards intelligence is related to the development of artificial intelligence. A deep relationship.

In 2018, artificial intelligence will be able to get rid of the dilemma of insufficient data in 2017, and enhance the ability of algorithm. It will promote breakthroughs in technologies such as voice and image recognition, and will further increase the added value of smart terminals and open up more intelligent elements. application. Retail services can also benefit from the evolution of artificial intelligence, analyze the attributes of customers to develop services that better meet customer needs, and will also increase the demand for intelligent terminal devices with audio and video and sensing.

In addition, artificial intelligence technology can also make the security defense system more complete. Hong Chunhui believes that the use of artificial intelligence enhancement capabilities from the outside of the system for emotional analysis, endpoint behavior anomaly analysis, and identity authentication will create more efficient and immediate security protection. And with the rapid increase in the number of Internet of Things (IoT) devices, the development of artificial intelligence applications for systems and products will become the main driving force for the artificial intelligence of the security system.