Compared to a few years ago, the AI market is starting to solidify around real-world applications with the pace of change being faster than it has ever been before, as startups and technology providers rush to create platforms and targeted niche solutions for solving specific enterprise problems. The industry is churning and evolving quickly as merger and acquisition (M&A) activities abound, and it is homing in on areas of focus. Tractica's 2018 AI ecosystem report covers a wide range of companies in the growing AI space, drawing on Tractica's internal knowledge base of the competitive environment, as well as external sources. Tractica has identified and categorized a total of 1,000 AI companies and provided more in-depth profiles of 200 key industry players that our analysis has revealed to be the most notable and representative examples of AI technology providers, solution providers, platforms, service providers, hardware vendors, and other players who each in their own way are helping to propel the sector forward. These companies span the globe and cover a spectrum of technologies and end-market segments.
In the past year MIFID II has enticed change and development across the financial markets and research sector. Here Fabrice Bouland, CEO of Alphametry, analyses said change and the impact it has had on innovation. Six months in and MiFID II research unbundling regulation has appeared to create an even worse market for investment research than we had previously. With many commentators decrying the'unintended consequences' of the new legislation – namely bringing the research market to a grinding halt as asset managers assess their needs and sparking a price war which has all but crippled smaller, niche research houses – one might wonder if there is anything positive to say about the impact of MiFID II on the research market and whether anything which can be done to revive it? In truth, MiFID II has ultimately shown us the historical ambiguity investment managers have always had with research.
When it comes to fraud detection, what a lot of people call AI is not really AI. Fraud systems that make algorithmic decisions, albeit complex, are not truly based in AI. Because the terminology is popular, many use the term for marketing purposes, with no real basis in the technology itself. AI is getting computers to act or do "smart" things that they are not explicitly programmed to do. In short, as John McCarthy – the scientist who coined the phrase "artificial intelligence" – said, AI is "making a machine behave in ways that would be called intelligent if a human were so behaving."
CB Insights is putting the numbers behind what industry insiders have already been noticing: Artificial intelligence is hot in healthcare right now. From a provider standpoint, many are just beginning to explore the possibilities and see how such capabilities can fit into the care delivery setting. Many providers are looking into patient readmissions as one area for a use case. However, due to the infancy of the current clinical use cases, artificial intelligence receives a fair amount of skepticism in the healthcare space. For one, "artificial intelligence" has become a catch-all shorthand for some disparate topics such as predictive analytics and machine learning.
Over 15 million artificial intelligence enabled devices will have been installed by the global industrial manufacturing sector in five years' time according to the latest forecast from technology market specialist ABI Research. AI is expected to be leveraged at different phases of the manufacturing process, from generative design in product development to machine vision, defect inspection, production optimization and predictive maintenance in the production phase. ABI Research predicts that the total installed base of AI enabled devices in industrial manufacturing will reach 15.4 million in 2024, with a compound annual growth rate (CAGR) of 64.8% from 2019 to 2024. "AI in industrial manufacturing is a story of edge implementation," said the firm's principal analyst Lian Jye Su. "Since manufacturers are not comfortable having their data transferred to a public cloud, nearly all industrial AI training and inference workloads happen at the edge, namely on device, gateways and on premise servers." Despite emerging AI enabled solutions and the wealth of data in the manufacturing environment, the implementation of AI in industrial manufacturing has not been as seamless as was expected by the industry, the research adds.