The 8th Overseas Talent Entrepreneurship Conference (OTEC), Data Intelligence and New Infrastructure Track, was successfully held as part of EqualOcean's flagship World Innovators Meet (WIM) Salon.
The 8th Overseas Talent Entrepreneurship Conference (OTEC), Data Intelligence and New Infrastructure Track, was successfully held as part of EqualOcean's flagship World Innovators Meet (WIM) Salon. The three prizewinners who most thoroughly impressed the judges and audience alike were Beijing BosonQ Technology, Panda Technology and Ningbo OE&N Safety and Science Technology; they claimed the first, second and third places.
The three champions also encompassed the technological development that China has been actively fostering: the pursuit of scientific breakthroughs, fintech export acceleration and manufacturing upgrades.
To gain a more comprehensive understanding of our three prizewinners and their work, EqualOcean conducted an in-depth interview with the founders to find out what made them excel and claim the OTEC top prizes.
As the top elite of OTEC, the founder of BosonQ Technology, Mr. Kai Wen expressed how "it is an honor for BosonQ Technology to win the OTEC Data Intelligence and New Infrastructure track. I am grateful for the Chaoyang Government leaders and the OTEC Committee for their support. While this victory plants a stepping stone to our commercialization, I hope together with the other outstanding projects; we can represent OTEC in carrying forward the mission to achieving greater success."
Mr. Xiaozhou Gao, Panda Technology's founder, also spoke highly of OTEC's organization "the whole event was very well organized and hosted, the organizers were enthusiastic and friendly, and the roadshow format was compact and suitable. All in all, it was a highly efficient and professional event. Thank you very much to all for giving us the opportunity and recognizing our business model."
Mr. Kai Tang, founder of OE&N commented, "I am thoroughly impressed with the quality of the participating projects and the professionalism manifested by the organizing body. Whether it is regarding the prizes or the follow-up procedures, the Chaoyang Government has been very supportive and generous. I have personally learned a lot from this experience and look forward to keeping further contact with the fellow contestants and organizers."
When discussing about BosonQ Technology's establishment, Mr. Kai Wen mentioned, "we are still in the early stages and are selecting a base; Mr. Xin Liu from Shanshu Technology, an OTEC-incubation company, had suggested to me that OTEC would be a trustworthy option, as they provide a full-package service for those seeking to start up a business. Therefore, if there is an opportunity, we very much hope to establish our AI co-processor CQX laboratory here in Chaoyang District."
Mr. Xiaozhou Gao from Panda Technology spoke about the Chaoyang District with full of memories, "I used to work at a big internet company, which was located in Chaoyang; thus, I am very familiar with the area. I am actually a local Beijinger. If there comes an opportunity, I will definitely consider starting a compnay in Chaoyang."
OE&N's Mr. Kai Tang said, "we previously established a subsidiary in Haidian District, if relevant policies are appropriate, we will consider relocating from Haidian to Chaoyang. In this process, we hope to get support from the Chaoyang District Government and EqualOcean."
Quantum technology is considered to be a future part of the fourth industrial revolution, and it holds a strategic and highly influential position under the current Sino-US tension. Quantum information mainly covers two fields: quantum communication and quantum computing. In the former field, China holds a leading position globally; however, China has achieved good progress in the latter field, though it is still catching up to the US. For example, Google was the first to release a 53-qubit quantum chip in September 2019, much in advancement compared to here in China. Regarding quantum computing, countries have been augmenting their investments in the past 20 years, deeming it difficult for any player to overtake the progress. This is due to the leading American technology enterprises such as Google and IBM having invested in quantum computing for more than a decade, along with their capital and labor investment that is beyond measures.
Mr. Kai Wen put across the chip manufacturing progress in the following analogy: "if the basis of calculation cannot circumvent Moore's Law, then China's new technology infrastructure will remain in a state of catching up."
Currently, Google's universal-use quantum computing processor solution has only achieved 50+ qubits, and it is expected to exceed 100 qubits in the next three years. This suggests that the commercialization of the universal quantum computing processor is still a long way off. On quantum computing processors achieving large-scale qubit calculations, Mr. Kai Wen commented, "companies such as Google and IBM use superconducting quantum computing solutions that need to be completed in a low-temperature vacuum environment. Therefore, increasing the scale of qubit calculation is undoubtedly a huge challenge; every increase requires massive investment and faces huge engineering difficulty." Therefore, during his Ph.D. at Standard University, Mr. Kai Wen proposed an alternative quantum AI co-processor CQX solution, "the quantum AI co-processor CQX solution uses coherent light pulse phase qubits, and the pulse is inputted through the controller. It has three leading technological features: qubit stability, quantum operation stability, and quantum state stability. Therefore, this solution can complete calculations at room temperature and in optical fibers, thereby achieving tens of thousands of qubit calculations fasters." Regarding the commercialization aspect, Mr. Kai Wen is confident: "at present, BosonQ has completed 1,000+ quantum computing prototypes and 10,000+ qubit calculation experiments, we will continue to develop prototypes with greater acceleration capacity. To achieve 100,000 or one million bits of computing, it is necessary to further improve the control accuracy and stability, as well as adopt a more mature commercialization strategy. In the long-term, we aim to achieve 100,000+ qubit quantum AI co-processor calculations within two to three years."
Compared with the already existing universal-use quantum computing processor solution, BosonQ Technology's AI co-processor CQX solution is more targeted and mainly used for artificial intelligence algorithm acceleration. Currently, the leading products in this field are NVIDIA's GPU and Google's TPU; Mr. Kai Wen elaborates on the difference between the quantum AI co-processor CQX solution and the NVIDIA solution. "NVIDIA accelerates the classic neutral network; network capacity increases in a linear fashion with the number of nodes and connections; thus, it is necessary to implement a better performing network, which involves a network upgrade that expands exponentially. BosonQ Technology uses quantum neural networks to achieve exponential acceleration and thus solve the bottleneck of classical network capacity and computing-bit requirements. This way, we would be able to acquire commercialization opportunities faster and attract investment in our technology." Mr. Kai Wen also revealed, "as far as I know, well-known foreign companies and institutions are also exploring towards this direction, which urges us to act fast to seize the opportunity. For example, we participated in the OTEC this year also with the hope of accelerating commercialization and increasing publicity."
In terms of entrepreneurship, the founder of Panda Technology, Mr. Xiaozhou Gao, shared his insight: "entrepreneurship is like a marathon, we are not expecting immediate results; instead, we are in it for the long game and build our competitiveness in the long-term."
Mr. Xiaozhou Gao initially started noticing fintech infrastructure opportunities in developing countries in 2017; he recalled, "I remember the first time I visited the region, what really stuck was the underdevelopment of local technical facilities, whether it be transportation, education, or more specifically, online payment. For instance, the degree of paperless currency for purchasing things in convenience stores was relatively low. However, in recent years, with the development of regional economy, it is clear that the mobile payment penetration rate has risen up significantly. For these reasons, third-party payment companies have gradually emerged in the area. Therefore, in 2018, we discovered this market's maturity and seized this opportunity to start our adventure there. All in all, with the improvement of the local payment infrastructure, alongside strengthening economic development and improving public affairs, we decided to fully dive into this market." Fast-forward to today, Panda Technology has not only obtained dozens of computer software copyrights and patents, but even more impressive is that the company has also accuired finance licenses and operating qualifications in countries such as Indonesia, Vietnam and the Philippines.
Facing the other fintech giants also extending their scope into the overseas market, Mr. Xiaozhou Gao is confident , "compared to our competitors, we hold two key advantages that make us excel – our fast decision-making and international experience."
Mr. Xiaozhou Gao revealed, "working in emerging markets requires our team to stay sharp and keep ourselves up to date with the latest market trend; Moreover, Panda Technology's organizational structure is relatively flat, our executives are working at the frontline whichenables the local team to be more flexible and decisive in the face of sudden market changes."
Regarding having international experience, Mr. Xiaozhou elaborates, "I think we have a pretty good international background with coreteam members having studied or worked in Europe, America and the Asia Pacific before. We also have gathered strong local resources in the South East Asia, which enhances our collaboration with our foreign businesspartners."
Finally, Mr. Xiaozhou Gao expressed that he is not too worried about pandemic, "with the continuous development of vaccines, it will gradually become under control, and the economy in these countrieswill recover.
Every year, tens of thousands of people die from safety accidents in China, with an even higher figure for the number of serious injuries. This results in a direct economic loss that exceeds CNY 100 billion. For instance, the coal industry's death rate is ten times that of the United States and 20 times that of Australia. Among these cases, 96% of personal injury accidents are directly caused by unsafe behavior. Mr. Kai Tang of OE&N expressed deep concern for this and an urgent need to solve the Chinese traditional industries' safety issues. Therefore, OE&N was founded to develop an "AI-based security intelligence platform," aiming to reduce such incidents.
In addition to the hundreds of billions of market demand, Mr. Kai Tang is also eager to show what he has learned from the work of the Nuclear Power Institute – he believes that in the era of the Internet of Everything, steadfast reform in security will be underway, which can be summarized as a 'safe and intelligent 3+2 plan,' which consists of three changes and two concepts. He also emphasized that, "as we all know, nuclear power safety is paramount, and it is no small matter." Nuclear power safety issues rarely occur because its philosophy differs from that of the traditional industries – allow an individual to make mistakes, but the organization itself needs to achieve zero accidents; through the constant improvement of the organization itself, thereby avoiding individual mistakes. These two concepts of nuclear power safety have also been implemented at OE&N as a fundamental guiding ideology for its product development.
Mr. Kai Tang explained, "according to relevant survey statistics, people make mistakes up to 100 times a day on average, and they are impossible to avoid. Therefore, when people make mistakes, the most important thing is to reduce or even eliminate accidents by using a safety framework. Next, is to redesign current safety architecture to reduce the occurrence of similar incidents or mistakes in the future. This is surely more effective than punishing your employees."
To materialize the two concepts, Mr. Kai Tang gave an example, "the reason why OE&N was able to defeat well-known Chinese manufacturers such as Huawei and New H3C Technologies, was evidently not because of our artificial intelligence technology, they are surely not weaker than us in terms of AI power; rather, it is the concept. When your opponent is still rooted in the traditional concept, they have already lost the competition. This is not the type of efficiency that solely AI algorithms can bring, but rather, it is about the conception."
Regarding OE&N's products, the 'AI-based security and intelligence platform' makes the following three transformations: from professionals in charge of security to everyone in charge of security, from manually controlled security to intelligence-based security; from on-site management of safety to switchboard controlled. "Traditional management is the supervision of a professional and knowledgeable safety direction randomly selected on the spot," says Mr. Kai Tang, "at present, OE&N's products are mainly used by relatively traditional large manufacturers. Including in electric power, thermal power, nuclear power, chemicals, machinery, and construction, and will continue to expand in the future. In these large factors, the area is huge, while only minimal management personnel can oversee safety, sometimes it is even impossible to find out the exact cause of an accident." On the other hand, OE&N uses the combination of high-tech software and hardware such as AI to help operators manage safety in real-time by making AI-powered judgment. In this way, the transformation from manual safety management to an intelligence-based solution has been formed.