The author personally tends to think that it is at the time of the Internet in 1995. It was the time when Ma Yun went to the United States to see the Internet excitedly and could not stand it. There are three differences: First, the center of the then Internet revolution was The United States, this smart revolution, China may play a role of "an important member of the core leadership." Second, the Internet took decades to accomplish today's achievements, and the pace of development of artificial intelligence may be much faster than the Internet, and it may be 20 years in 5-10 years. Third, the changes that artificial intelligence brings to human society will be bigger than the Internet, and the scale of business will be much larger. According to Li Kaifu's words, it may be thousands of times. In addition to the above three points, the author believes that the current stage of industrial development of artificial intelligence is equivalent to the Internet in 1995 - the beginning of mature technology, basic supply of capital, public interest index amplification, and a serious shortage of industrial applications. Based on this understanding, I feel that for any company and ordinary people, the opportunities are great. Since Yahoo, Netscape, Microsoft, Cisco, etc. were available in 1995, BAT, Facebook, JD.com, Xiaomi, 360... How to grasp the wave of artificial intelligence? The opportunity is really great and the time is also very good, but the time that we left for us is not that long. In a few years time, there will be a high probability that BAT in the era of artificial intelligence will emerge. So, how do non-AI companies meet such a big deal? In the next few years, the five types of artificial intelligence in the "service intelligence" era will be: Mode 1: Ecological Builders - The whole industrial chain ecology + scene application as a breakthrough. Mode 2: The driver of the technology algorithm - the technology layer + scene application as a breakthrough. Mode 3: Application Focuser - Scene Application. Pattern four: vertical field pioneer - killer application + gradually build vertical field ecology. Pattern 5: Infrastructure providers - cut in from the infrastructure and expand downstream of the industry chain. How do non-AI companies catch up with the wave of artificial intelligence? Then, how are the above five models selected? We categorize non-AI companies (and large companies that do not already include AI, such as BAT, who will take the first, second, or fifth model) into roughly three categories: The first category: online and High-tech-based large companies (Unicorn and above) These companies have good resources and fame, and they are much more attractive to talents, resources, and funds than the other two categories. But because the AI wave is really too fierce, some companies will worry about their own resources are not much, can not put too much effort in AI, but are very worried about missing this wave, the cost will be greater to catch up. For this concern, we must first consider a question, and make it clear whether AI really has a huge positive impact on your business. Is it the key to the success or failure of your business? If the answer is yes, then try hard to do AI. Note that AI is to be done, not to look at it, to learn it, and to do it quickly is better than others. Note that it's not too late to do it right now. The current AI-related resources open source is a big trend. The above application based on open source software is the most cost-effective AI solution. As mentioned above, such companies are also somewhat famous for their money, so it is advisable to set up their own AI team. This will help to establish the threshold and the efficiency of applying AI technology will be much higher. The second category: Small businesses with online and high technology as the main body (cannot reach the size of unicorns) The smell of these companies is very sensitive, but money, people and resources are relatively scarce. Like the first type of enterprises, Find out if AI is a positive factor in winning the war to dominate the industry and decide to do it well. You can take the third or fourth model. However, because it is not so financially rich, you can directly use the AI service provided by the giants on the cloud, and consider the big company as its own infrastructure. The third category: Non-technical companies with offline as their main body As we all know, artificial intelligence requires the combination of “algorithm+hardware+data”. Among these three factors, this category of enterprises holds data. Similarly, if you ask the above question, if the answer is no, then it will be more practical to make the existing business bigger and stronger, and it will not be a big issue to take another AI strategy in another three or five years (but it is also recommended that you pay attention to the trends of AI, say Indefinitely, the demand for AI will come out one day.) If you want to understand the wave of artificial intelligence in 2017, I suggest these points: First of all, most of these companies did not attach importance to the accumulation of data before. Although there is no time and place, it must be followed. Secondly, the third model was adopted in the second entrepreneurial mentality and in vitro incubation, and the scene was applied based on the above accumulated data. Further conditions mature to take the fourth mode. Finally, like the second-tier companies, don't build AI teams yourself. Use open-source software provided by big companies and cloud AI services. Such services will surely come out in 2017—because they lack data.
Plastic Cap,Airless Pump Bottle,Disc Top Cap
Ruichang Commodity Packaging Co., Ltd. , http://www.nbcosmeticpackaging.com
没有评论:
发表评论