The Industrial Revolution of the early 19th century saw humans turning to machines to take on human capabilities in order to increase productivity, enable the development of new products, and transform the business model. Now in the 21st, humans are yet again turning to machines, specifically artificial intelligence and machine learning, as a means to increase productivity, offer new products and services, and transform the business model. A 2019 Gartner CIO survey found that while only 14% of organizations currently employ AI, nearly 50% intend to do so by 2020[1]. A primary hurdle to overcome in order to benefit and deliver a return on investment from this modern day revolution, is the development of an effective AI strategy, including knowing what tools to use to harness the operational use of AI, and bring digital transformation to enterprise. Despite a strong desire to integrate AI, recent surveys reveal that top CIO challenges include: defining an AI strategy; identifying use cases; and confusion over enterprise AI application types and vendors in the marketplace. The question therefore becomes, how can enterprise harness and exploit this revolutionary opportunity?
Understand AI. First, understand that AI is a game changing technology for the use of data and analytics. AI applies advanced analysis and logic-based techniques, including machine learning, to interpret data and events and to automate decision-making and action. Often broken into narrow vs. general intelligence, AI refers to a system that is capable of demonstrating forms of behavior that mimic human intelligence. Machine learning is a subset of AI and is generally divided into supervised vs. unsupervised learning. The former involves teaching a system how to respond to data, by example. For instance, teaching a system how to identify spam mail so that once it is taught the features of spam, it will automatically apply this information to new data, identifying and removing spam. By contrast, unsupervised learning uses algorithms to identify patterns in data, looking for similarities that can be used to identify features and create labels, without having to be specifically programmed to do that task. An in-depth technical dive on AI, ML and concepts like neural networks and deep learning are beyond the scope of this article, but a CIO needs to firmly understand how AI works so that he or she can operationalize AI into a digital business model.
Understand Opportunity. Next, understand the AI use cases in your industry. For example, everyone understands that Siri recognizes and responds to your voice, that Google stalks your every search to bombard you with ads, and not to answer the home phone because its a computer calling to see if your air ducts need cleaning. Early users of AI applications, such as Amazon and Netflix, recommend products and movies based on your previous selections; your past choices are incredibly valuable data. But what does AI mean for businesses such as banking and investment, healthcare, insurance, manufacturing, natural resources, and utilities. It is imperative to understand how AI can solve business problems. All of these businesses, regardless of industry, have a common goal and seek to achieve greater agility, better customer experiences, cost savings, and revenue growth — all of which can be realized by embracing a new digital business model that incorporates artificial intelligence.
A recent survey regarding organizational plans for the use of enterprise AI applications, found that thirty-seven (37%) of those in banking and investment services had deployed AI or planned to do so within 12 months[2]. Top three uses in banking were reported as fraud analysis on transactional data, chatbots, and market/consumer segmentation. These companies find success by leveraging technology to revolutionize finance. Leveraging machine learning to detect fraudulent and abnormal financial behavior helps lower operational costs. Other fintech enterprise applications use machine learning to improve basic business accounting, including expense reporting, billing, AI for credit lending, AI techniques for financial analysis, trading and regulatory compliance. There are even AI chat bots aka robo-advisory who tell you what to do with your money, rather than humans. All enterprise AI use cases are intended to lower operating and compliance costs, reduce human error and improve efficiency by incorporating a new digital business model.
A whopping forty-eight per cent (48%) of those surveyed in the insurance industry reported they had enterprise AI deployed or planned to do so within the next 12 months[3]. Top uses for AI in the insurance industry include fraud analysis, chatbots and process optimization tools. Process optimization refers to those types of enterprise applications that leverage machine learning to do traditional business operations such as quote optimal premium pricing, manage insured claims effectively, and improve overall customer satisfaction through real time access to data and claims information. These insurance industry enterprise AI apps reduce operational costs, while at the same time deliver meaningful customer experiences.
Thirty-eight per cent (38%) of healthcare providers surveyed had deployed AI or planned to do so within the next 12 months[4]. The top use cases in healthcare include computer diagnostics, process optimization tools, and market/consumer segmentation. Healthcare AI use cases span the spectrum – from fascinating improvements in diagnosis and treatment to mundane claims processing. Patient and treatment data analytics are used to discover insights, suggest actions, find the best treatment plans, find new drugs, and even perform prescriptive analytics to enable real time case prioritization and triage. AI-related technologies such as IBM’s Watson clinical decision support tool are being found in hospitals all over the world. Leveraging enterprise AI in healthcare improves mortality rate and patient satisfaction while reducing healthcare costs. The bottom line, knowing how to use AI in healthcare saves lives.
Understand and invest in the right technology. AI tech includes a broad range of AI enterprise applications as well as ones enterprise can build themselves. According to a recent survey, at least 17% of new apps in 2019, will involve AI or some machine learning functionality[5]. Microsoft recently announced a new tool that provides app makers the ability to add AI capabilities to their apps in order to create more intelligent applications. Similarly, Crowd Machine’s Crowd App Studio, where app makers build enterprise applications without writing any code, also offers this capability, but takes it a step further. A Crowd Machine developer can integrate AI into new custom flexible scalable no code enterprise grade applications, but can also decompose out-dated legacy systems into functional microservices or components. These functional components can include AI bots and machine learning capabilities to intelligently access, utilize and provide real time visualization of previously trapped legacy data, or data from multiple disparate sources. Data analytics on AI steroids. Having the ability to understand what tools exist, and which tools provide the greatest level of flexibility to ensure continuous digital agility is an essential component of the AI strategy for enterprise.
CEOs and CIOs believe that AI will be strategically important to business, but face many challenges in the process of implementing an AI strategy and the tools that will guarantee an ROI. Investing in AI solutions and which types of enterprise applications will depend on your industry and use cases available. While many businesses rely on chatbots, it doesn’t mean your business needs to use this common AI app. As Microsoft and Crowd Machine recognize, the ability to rapidly leverage AI with flexible custom scalable no code enterprise applications and functional components could be the most effective means of implementing a successful AI strategy. Leveraging AI with an effective strategy and the best tools provides enterprise the ability to stay agile, scale, and maintain continuous adaptive environments with sustainable competitive advantage.
[1] “Artificial Intelligence Primer for 2019,” ID G00709933, Jan. 17, 2019; Whit Andrews
[2] Id.
[3] Id.
[4} Id.
[5] The State of Application Development, 2019/2020, Outsystems.