Introduction<\/h1>\n
In this post, I discuss the development of the Enterprise AI business case<\/strong> through a framework of four quadrants. \u00a0According to Gartner: \u201cThe mindset shift required for AI can lead to \u201ccultural anxiety\u201d because it calls for a deep change in behaviors and ways of thinking\u201d.<\/em> Deployment of AI in an Enterprise is complex and multi-disciplinary. Hence, this framework is evolutionary. \u00a0The vendors and initiatives listed are included to illustrate the framework.<\/p>\n In progressive orders of complexity (and opportunity) the four quadrants for the Enterprise AI business case<\/strong> are:<\/p>\n The analysis is based on the Enterprise AI workshop in London and remotely<\/a><\/p>\n \u00a0 Enterprise AI is an abstract concept, interdisciplinary and much-hyped concept. But in any case, the deployment of AI in the Enterprise cannot be viewed in isolation<\/strong>. Within the Enterprise, there already exist systems (like ERP and Data Warehousing). The integration of these will have a role to play in any AI deployment. The word \u2018Enterprise\u2019 can be seen in terms of Enterprise workflows<\/strong>. We also consider the core Enterprise (a non-manufacturing company ex Insurance) and the Wider enterprise (including supply chain). Hence, Enterprise AI could be understanding how workflows change when AI is deployed in the Enterprise.\u00a0<\/p>\n The professional deployment of AI in Enterprises differs from the content in a typical training course<\/strong>. In larger organisations, the Data Science function typically spans three distinct roles: The Data Engineer, the Data Scientist and the DevOps Engineer<\/strong>. The Data Scientist is primarily responsible for developing the Machine Learning and Deep Learning algorithms. The Data Engineer and The DevOps Engineer roles work in conjunction with the Data Scientist to manage the product\/service lifecycle. Hence, in an Enterprise, managing the AI pipeline involves the philosophy of CICD (Continuous Improvement \u2013 Continuous Delivery). CI\/CD can be seen as an evolution of Waterfall and Agile methodologies.<\/p>\n Finally, to clarify some definitions used<\/strong>: Machine Learning<\/strong>: Means systems that can learn from experience (Data); Deep Learning<\/strong>: Implies a system that can perform automatic feature detection based on Deep neural networks; Artificial Intelligence<\/strong> involves machines that can reason.<\/p>\n<\/p>\n With this background, let us explore the four quadrants of the AI business case<\/p>\n We could initially model the problem as a machine learning or a deep learning problem. At this stage, we are concerned with the accuracy, choice and the efficiency of the model. \u00a0Hence, the first quadrant is characterized by experimental analysis to prove value.<\/p>\n We are also concerned with improving the existing KPIs. For example, if you are working with fraud detection or loan prediction \u2013 each of these applications has an existing KPI based on current techniques. The machine learning and deep learning models would be expected to significantly improve the current benchmarks. We are typically working with one node(non-distributed) processing. The Data could be in Time series, Tabular, Textual, Image, Audio or Video based. The applications could involve Computer vision, NLP, Fintech\/financial services, Healthcare, Reinforcement learning, Unsupervised learning (ex GANs, VAE), Emotion AI (Affective computing) etc. Deep learning architectures<\/a> are rapidly evolving. \u00a0Hence, there is a lot of effort and skill needed at this stage.<\/p>\n Building on from the first quadrant, the second quadrant is characterized by<\/p>\n Both ERP and Data Warehousing exist in large Enterprises. Apart from integration with existing system and with a Cloud strategy, in this quadrant we have to also consider<\/p>\n Marlene Jia creates a landscape of Enterprise AI companies<\/a> which categorizes AI applications in the Enterprise. \u00a0We note that the problems are the same or similar as before but are solved more optimally using AI by gaining insights from much larger amounts of (often) unstructured data.\u00a0 The categories include BUSINESS INTELLIGENCE ex Ayasdi; PRODUCTIVITY ex virtual scheduling assistants like X.ai; CUSTOMER MANAGEMENT ex Inbenta\u2019s AI-powered natural language search; HR & TALENT Companies ex Entelo; B2B SALES & MARKETING Salesforce\u2019s Einstein; CONSUMER MARKETING ex companies like Lexalytics; FINANCE & OPERATIONS ex AppZen which is an automated audit platform that can instantly detect fraud and compliance issues; DIGITAL COMMERCE ex Sentient Technologies analyzes product recommendations for user actions; DATA SCIENCE like RapidMiner; ENGINEERING companies like Diffbot; \u00a0SECURITY & RISK ex Demisto (incident response); INDUSTRIALS & MANUFACTURING ex GE Predix<\/p>\n The above analysis also demonstrates that AI will impact many areas of the Enterprise but in this Quadrant, the emphasis is on evolution rather than revolution<\/strong> where companies integrate with existing applications and also gain experience in AI. The challenges in this quadrant are mostly data related \u2013 especially the challenges of finding labelled data.<\/p>\n In the third quadrant, the emphasis is scaling and in handling real-time transactions<\/strong>. There are a range of technologies which may be involved here \u2013 which mostly come under the category of CICD also a range of initiatives from Enterprise Cloud providers like Azure ML CI\/CD<\/a><\/p>\n At a simplest level, we can deploy deep learning models using flask<\/a> but more complex strategies could come into play for example Mlflow from databricks<\/a> , kafka to take on some functions of Enterprise service bus<\/a>, use of Jenkins 2.0 AI pipeline models for continuous delivery<\/a> and others<\/p>\n \u00a0<\/p>\n Quadrant four is the most interesting. It is driven by AI talent who can think strategically and technically<\/strong>. At this stage, we are looking for AI already integrated into the Enterprise and how AI can be used for disruption. This calls for an AI first company<\/a> as outlined by William Vorhies and four major AI strategies<\/a>\u00a0 for AI i.e. Data Dominance, Horizontal, Vertical and Systems of Intelligence. Monica Rogati also talks of an AI hierarchy of needs<\/a> which also resonates with this approach.<\/p>\n In this quadrant, we are working with issues like Process alignment with AI at the core, IPR, AI and humans working together, Cobots etc. The work is driven mostly by Masters\/PhD often implementing ideas directly from research papers to create new IP. This quadrant also involves a change in company culture and role of people within a company<\/p>\n \u00a0\u00a0<\/p>\n In this post, we discussed the development of the Enterprise AI business case<\/strong> through a framework of four quadrants. \u00a0The disruption from AI for the Enterprise will be evolutionary. The first two quadrants are based on incremental changes. But the radical disruption is very real. Hence, the last quadrant involves both disruption and stagnation. AI is a winner takes all game. An alternate working definition of AI is \u2018processes which improves with experience\u2019. In that sense, the early adopters who will learn from AI will be the future market disruptors. \u00a0<\/p>\n The analysis is based on the Enterprise AI workshop in London and remotely<\/a><\/p>\n Welcome your comments to ajit.jaokar at gmail.com<\/p>\n \u00a0<\/p>\n<\/div>\n Go to Source<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"\n
![](\"http:\/\/api.ning.com\/files\/VOs33rkZCSAV5*BYSG6vFe1s39lftSomhB7vXB7qYeR-w6v9tD1fI-T7g7pDe0IsPr9sLJzsMEiImPz4itXMxOYxUDEXm8zs\/fourquadrantsoftheEnterpriseAIbusinesscase.jpg\")
<\/a>\u00a0
Background<\/span>\u00a0<\/p>\nEnterprise AI Business Case<\/h1>\n
Machine Learning and Deep Learning<\/h2>\n
Enterprise Platforms and Data<\/h2>\n
\n
\n
AI Pipeline and Scalability<\/h2>\n
AI disruption (and Stagnation)<\/h2>\n
Conclusions<\/h1>\n