.
Presenters: Howie Xu (Founder & CEO, TrustPath), Joy Tang (Founder & CEO, Markable.AI), Terry Song (Managing Director, JD Capital), Julie Choi (Head of AI Marketing, Intel), Shaun Paga (VP of Sales at Soul Machines), Anis Uzzaman (General Partner, Fenox Venture Capital)
To read the full report click here for the digital edition.
Enterprises have increasingly adopted AI. This has led to a diverse artificial intelligence ecosystem around the world, involving not only large businesses, but AI startups and venture capitalists as well. Although the different AI players often face dissimilar challenges and opportunities, they have increasingly incorporated artificial intelligence into their operations to capitalize on its many benefits.
The “AI in Enterprise” panel gathered speakers from Fenox Venture Capital, Markable.AI, TrustPath, Soul Machines, Intel and JD. The speakers applied their professional expertise and international perspectives to discuss AI’s current role—and predict its future role—in enterprise. They highlighted the complementary nature of China and the United States’ AI entrepreneurial environments and asserted that AI will continue to drive companies and societies into the future.
KEY TAKEAWAYS
Startups and large enterprises face different opportunities and challenges in AI. Enterprises and startups occupy different roles in the economy. As companies’ scale, scope and time in business differ, so do their strengths and weaknesses. Although artificial intelligence has increased competition between startups and enterprises, with both being AI-focused in applying its technology into their businesses, AI implementation has led startups and companies to witness different opportunities and hurdles.
AI startups benefit from their focus and collaborations. AI companies’ narrow focus makes them more agile in the changing market. Joy Tang notes the importance of being niche in a sector, highlighting the narrowness of fashion recognition and its accompanying technological difficulty as an example. Being niche creates accuracy, which enables AI companies to become more advanced as they focus on solving a single problem well. To help solve problems, AI startups develop symbiotic relationships with small, medium and large enterprise clients: smaller AI startups give their models—their software development kit (SDK) and underlying application program interface (API)—in exchange for the enterprise’s data. Software companies like Markable. AI thrive in this AI ecosystem by shortening enterprises’ technology development time while receiving needed and diverse data. Since data is “the blood source” of AI, such relationships are vital.
Large companies see benefits in efficiency improvements. One of companies’ greatest AI opportunities is improving logistics and operations. Specifically, Terry Song highlights JD’s “4 Deployment of Inventories” strategy as a way to improve customer experiences by cutting down delivery times. Rather than having one large inventory warehouse at the outskirt of urban areas, this approach localizes travel time by deploying 4 separate inventory warehouses throughout the city. By cutting down on transportation times, consumers can receive orders within the same day.
Enterprises struggle with AI’s short market window. Since AI is fast-paced and disruptive, products fall in and out of demand. By the time an enterprise recognizes a key industry or product and decides to hire people and build a team, two or three years have passed and the company has missed the market window. Companies can overcome this challenge by finding startups with a very specialized company product and teaming together.
AI companies face challenges with worker pay. Startups’ largest AI struggle relates to paying its workers. While startups are at risk of losing their shares to investors to keep fundraising, the employed engineers, data scientists and developers require adequate pay since they are highly qualified and essential to the company’s existence. As there are talent and qualification scarcities in AI, startups need to make competitive offers.
Artificial intelligence talent is scarce. Artificial intelligence requires an advanced skill set. Because of this, companies need qualified and educated data scientists, developers and engineers to develop and operate artificial intelligence technology. As AI bleeds into enterprises in terms of developing AI products and using AI to make more efficient and automate company processes, businesses compete against each other for talented and qualified employees— making talent recruitment and retention a pressing topic in regards to AI in enterprise.
Universities are essential to recruitment. There are four main universities where AI talent originates: Stanford, Carnegie Mellon, Massachusetts Institute of Technology (MIT) and University of California, Berkeley. Anis Uzzaman and Julie Choi recognize that AI talent does not come from elsewhere, with Chinese computer scientists attending these universities after their bachelor degrees as well. Because of this, AI startups and various enterprises go to universities when seeking talent. Tang specifically found talent by reading CVPR papers related to deformable object recognition and hired the students and professors that wrote the reports.
International universities are emerging as potential recruitment sources. Though the top four American AI universities definitely remain the top sources for AI talent, the University of Auckland, Shaun Paga highlights, plays a large role in providing Soul Machines with qualified employees. The company’s talent is not only trained in data and computer science, but have PhDs in diverse areas of psychology, neuroscience and neurology. Silicon Valley, Anis Uzzaman notes, is attempting to tap into Canadian markets in Toronto and Ontario because of the University of Toronto and McGill University’s noteworthy computer science programs. The narrow sources of AI talent, Choi asserts, has diminished creativity in AI. With more universities in New Zealand and Canada being viewed as talent sources, an AI “renaissance” could possibly occur.
Empowering AI talent retains them. As AI becomes more widespread, more enterprises seek qualified AI developers, engineers and computer and data scientists. To retain such workers startups and large enterprises offer different benefits. Startups, Tang asserts, give workers more ownership, control and creative licensing over technologies and products than any large firm. Having AI-trained and focused leadership is also a perk of startups: when AI talent is asked to create training data, AI’s fuel, workers are not pleased. Recognizing the importance of such data to the algorithm helps startups retain their employees.
China’s high company valuations and AI culture is attracting more talent. Although Chinese computer science students come to the United States for school and stay to work, Chinese AI workers are beginning to return to China to create AI companies. This is because of China’s friendly AI culture, a high growth rate, access to abundant data and Chinese AI companies’ high valuations—being valued three to five times more than their Bay Area counterparts. This, Tang notes, has caused Chinese AI talent to be less focused than American AI talent. As Chinese talent is funneled into large enterprises in China, their skills are at risk of becoming irrelevant as they do a single task in the large company but AI advances quickly in the outside world.
Large enterprises have acclimated to the AI ecosystem. Well established companies have adapted to AI technologies in order to reap artificial intelligence’s benefits. Even though large enterprises’ scale can inhibit quick adaptivity, these companies recognize the importance of AI in remaining relevant and competitive in today’s technological global economy. Large enterprises’ have focused their resources into applying AI to their business models, products and operations, bettering technology and their companies in the process.
Intel is more than hardware and CPU. Intel, the hardware company that invented processors, continues to play an important role in the tech industry. Intel has increasingly become an AI company as it provides the CPU and silicon needed to power artificial intelligence, with Microsoft’s use of Intel’s Field Programmable Gate Arrays (FPGAs) serving as an example. As AI models are the sum of data, compute and software, Intel strives to connect all parts of the equation. Specifically, the company is working with various organizations to create models in the form of PoCs and open sourcing tools for the data science community to accelerate enterprises’ AI development. Additionally, Intel is improving software data and usage to help enterprises maximize their CPU usage and quicken their models.
AI is central to JD’s business. The E-commerce company has adjusted its business over the past five years to make AI the core of its operations. Approximately 200 billion dollars of gross merchandise volume (GMV) was on JD’s platform in 2017. As that amount grows at an annual rate of 30 to 40 percent, the company recognizes it has reached a scale where it physically cannot do anything manually. This, coupled with rising labor costs, has pushed JD to use AI in fully automating the company’s warehouses in order to increase operating efficiency.
AI’s future depends on investment and human engagement. To fuel AI in enterprise, companies need money. As venture capitalists look for companies to invest in, they take an anticipatory approach to determine which companies will profit moving forward, as well as which products will remain relevant. Although there are many uncertainties in this line of work, technologies that improve the human experience and human-AI engagements will take center stage as artificial intelligence continues to disrupt enterprises and economies.
Future AI should prioritize trust and engagement. Soul Machines has made a new interface for AI. Although artificial intelligence pushes humans into more passive positions in many activities, this technology focuses on humans and human interaction, as they are the intended users. Soul Machines fully autonomous digital being, prioritizes trust and engagement when interacting with humans by detecting voice tones and facial expressions. By giving technologies a higher emotional quotient, AI will be more adopted in the future.
Venture capitalists are looking to automation and blockchain. AI’s fast paced and adaptable nature corresponds to venture capitalists’ forward thinking, but adds difficulty to VC’s line of work. The largest hurdle that venture capitalists face is anticipating what is next in making instantaneous investment decisions. Although there are many uncertainties, Uzzaman forecasts that automation in the robotic control space will be the next big thing to invest in. Blockchain also plays a role moving forward if coupled with AI correctly since AI is past experience and blockchain applies that experience to making it secure in the future. If a company can unite a blockchain-based distribution and data transfer with AI-based future analytics, they will receive investments.
The views presented in this article are the author’s own and do not necessarily represent the views of any other organization.
a global affairs media network
Report: AI in Enterprise
Artificial intelligence. Technology web background. Virtual concept
November 9, 2018
Presenters: Howie Xu (Founder & CEO, TrustPath), Joy Tang (Founder & CEO, Markable.AI), Terry Song (Managing Director, JD Capital), Julie Choi (Head of AI Marketing, Intel), Shaun Paga (VP of Sales at Soul Machines), Anis Uzzaman (General Partner, Fenox Venture Capital)
To read the full report click here for the digital edition.
Enterprises have increasingly adopted AI. This has led to a diverse artificial intelligence ecosystem around the world, involving not only large businesses, but AI startups and venture capitalists as well. Although the different AI players often face dissimilar challenges and opportunities, they have increasingly incorporated artificial intelligence into their operations to capitalize on its many benefits.
The “AI in Enterprise” panel gathered speakers from Fenox Venture Capital, Markable.AI, TrustPath, Soul Machines, Intel and JD. The speakers applied their professional expertise and international perspectives to discuss AI’s current role—and predict its future role—in enterprise. They highlighted the complementary nature of China and the United States’ AI entrepreneurial environments and asserted that AI will continue to drive companies and societies into the future.
KEY TAKEAWAYS
Startups and large enterprises face different opportunities and challenges in AI. Enterprises and startups occupy different roles in the economy. As companies’ scale, scope and time in business differ, so do their strengths and weaknesses. Although artificial intelligence has increased competition between startups and enterprises, with both being AI-focused in applying its technology into their businesses, AI implementation has led startups and companies to witness different opportunities and hurdles.
AI startups benefit from their focus and collaborations. AI companies’ narrow focus makes them more agile in the changing market. Joy Tang notes the importance of being niche in a sector, highlighting the narrowness of fashion recognition and its accompanying technological difficulty as an example. Being niche creates accuracy, which enables AI companies to become more advanced as they focus on solving a single problem well. To help solve problems, AI startups develop symbiotic relationships with small, medium and large enterprise clients: smaller AI startups give their models—their software development kit (SDK) and underlying application program interface (API)—in exchange for the enterprise’s data. Software companies like Markable. AI thrive in this AI ecosystem by shortening enterprises’ technology development time while receiving needed and diverse data. Since data is “the blood source” of AI, such relationships are vital.
Large companies see benefits in efficiency improvements. One of companies’ greatest AI opportunities is improving logistics and operations. Specifically, Terry Song highlights JD’s “4 Deployment of Inventories” strategy as a way to improve customer experiences by cutting down delivery times. Rather than having one large inventory warehouse at the outskirt of urban areas, this approach localizes travel time by deploying 4 separate inventory warehouses throughout the city. By cutting down on transportation times, consumers can receive orders within the same day.
Enterprises struggle with AI’s short market window. Since AI is fast-paced and disruptive, products fall in and out of demand. By the time an enterprise recognizes a key industry or product and decides to hire people and build a team, two or three years have passed and the company has missed the market window. Companies can overcome this challenge by finding startups with a very specialized company product and teaming together.
AI companies face challenges with worker pay. Startups’ largest AI struggle relates to paying its workers. While startups are at risk of losing their shares to investors to keep fundraising, the employed engineers, data scientists and developers require adequate pay since they are highly qualified and essential to the company’s existence. As there are talent and qualification scarcities in AI, startups need to make competitive offers.
Artificial intelligence talent is scarce. Artificial intelligence requires an advanced skill set. Because of this, companies need qualified and educated data scientists, developers and engineers to develop and operate artificial intelligence technology. As AI bleeds into enterprises in terms of developing AI products and using AI to make more efficient and automate company processes, businesses compete against each other for talented and qualified employees— making talent recruitment and retention a pressing topic in regards to AI in enterprise.
Universities are essential to recruitment. There are four main universities where AI talent originates: Stanford, Carnegie Mellon, Massachusetts Institute of Technology (MIT) and University of California, Berkeley. Anis Uzzaman and Julie Choi recognize that AI talent does not come from elsewhere, with Chinese computer scientists attending these universities after their bachelor degrees as well. Because of this, AI startups and various enterprises go to universities when seeking talent. Tang specifically found talent by reading CVPR papers related to deformable object recognition and hired the students and professors that wrote the reports.
International universities are emerging as potential recruitment sources. Though the top four American AI universities definitely remain the top sources for AI talent, the University of Auckland, Shaun Paga highlights, plays a large role in providing Soul Machines with qualified employees. The company’s talent is not only trained in data and computer science, but have PhDs in diverse areas of psychology, neuroscience and neurology. Silicon Valley, Anis Uzzaman notes, is attempting to tap into Canadian markets in Toronto and Ontario because of the University of Toronto and McGill University’s noteworthy computer science programs. The narrow sources of AI talent, Choi asserts, has diminished creativity in AI. With more universities in New Zealand and Canada being viewed as talent sources, an AI “renaissance” could possibly occur.
Empowering AI talent retains them. As AI becomes more widespread, more enterprises seek qualified AI developers, engineers and computer and data scientists. To retain such workers startups and large enterprises offer different benefits. Startups, Tang asserts, give workers more ownership, control and creative licensing over technologies and products than any large firm. Having AI-trained and focused leadership is also a perk of startups: when AI talent is asked to create training data, AI’s fuel, workers are not pleased. Recognizing the importance of such data to the algorithm helps startups retain their employees.
China’s high company valuations and AI culture is attracting more talent. Although Chinese computer science students come to the United States for school and stay to work, Chinese AI workers are beginning to return to China to create AI companies. This is because of China’s friendly AI culture, a high growth rate, access to abundant data and Chinese AI companies’ high valuations—being valued three to five times more than their Bay Area counterparts. This, Tang notes, has caused Chinese AI talent to be less focused than American AI talent. As Chinese talent is funneled into large enterprises in China, their skills are at risk of becoming irrelevant as they do a single task in the large company but AI advances quickly in the outside world.
Large enterprises have acclimated to the AI ecosystem. Well established companies have adapted to AI technologies in order to reap artificial intelligence’s benefits. Even though large enterprises’ scale can inhibit quick adaptivity, these companies recognize the importance of AI in remaining relevant and competitive in today’s technological global economy. Large enterprises’ have focused their resources into applying AI to their business models, products and operations, bettering technology and their companies in the process.
Intel is more than hardware and CPU. Intel, the hardware company that invented processors, continues to play an important role in the tech industry. Intel has increasingly become an AI company as it provides the CPU and silicon needed to power artificial intelligence, with Microsoft’s use of Intel’s Field Programmable Gate Arrays (FPGAs) serving as an example. As AI models are the sum of data, compute and software, Intel strives to connect all parts of the equation. Specifically, the company is working with various organizations to create models in the form of PoCs and open sourcing tools for the data science community to accelerate enterprises’ AI development. Additionally, Intel is improving software data and usage to help enterprises maximize their CPU usage and quicken their models.
AI is central to JD’s business. The E-commerce company has adjusted its business over the past five years to make AI the core of its operations. Approximately 200 billion dollars of gross merchandise volume (GMV) was on JD’s platform in 2017. As that amount grows at an annual rate of 30 to 40 percent, the company recognizes it has reached a scale where it physically cannot do anything manually. This, coupled with rising labor costs, has pushed JD to use AI in fully automating the company’s warehouses in order to increase operating efficiency.
AI’s future depends on investment and human engagement. To fuel AI in enterprise, companies need money. As venture capitalists look for companies to invest in, they take an anticipatory approach to determine which companies will profit moving forward, as well as which products will remain relevant. Although there are many uncertainties in this line of work, technologies that improve the human experience and human-AI engagements will take center stage as artificial intelligence continues to disrupt enterprises and economies.
Future AI should prioritize trust and engagement. Soul Machines has made a new interface for AI. Although artificial intelligence pushes humans into more passive positions in many activities, this technology focuses on humans and human interaction, as they are the intended users. Soul Machines fully autonomous digital being, prioritizes trust and engagement when interacting with humans by detecting voice tones and facial expressions. By giving technologies a higher emotional quotient, AI will be more adopted in the future.
Venture capitalists are looking to automation and blockchain. AI’s fast paced and adaptable nature corresponds to venture capitalists’ forward thinking, but adds difficulty to VC’s line of work. The largest hurdle that venture capitalists face is anticipating what is next in making instantaneous investment decisions. Although there are many uncertainties, Uzzaman forecasts that automation in the robotic control space will be the next big thing to invest in. Blockchain also plays a role moving forward if coupled with AI correctly since AI is past experience and blockchain applies that experience to making it secure in the future. If a company can unite a blockchain-based distribution and data transfer with AI-based future analytics, they will receive investments.
The views presented in this article are the author’s own and do not necessarily represent the views of any other organization.
