10 Analytics & AI Predictions for 2020
The year when AI goes from a buzzword to being understood as a valid business transformation.
ONE: COMPANIES WILL CONTINUE FAILING ANALYTICS & AI TRANSFORMATIONS
According to Harvard Business Review companies keep failing AI transformations. 77% of executives report that business adoption of Big Data/AI is a major challenge and 93% of respondents identify people and process issues as the obstacle. According to PWC, only 4% of executives surveyed plan to deploy AI enterprise-wide in 2020, versus nearly 20% a year ago. In a nutshell, 2020 is the year of the “reality check,” for companies who lay the groundwork for a realistic AI-powered transformation. AI will go from being a buzzword that everyone wants to try to be understood as a proper business transformation.
TWO: AI ETHICS STANDARDS WILL EMERGE
In 2020 ethical guidelines for AI will start to be standardized in business and society and some of them will shape regulations. The current General Data Protection Regulation (GDPR) focuses mainly on data security, privacy, and ownership. AI ethics standards will go far beyond this covering avoidance of unfair biases, explainability, and interpretability of AI systems (instead of black-box machine learning modules), human oversight of AI systems, etc.
THREE: AUGMENTED ANALYTICS TOOLS WILL BECOME MAINSTREAM
In 2020 companies will start using massively augmented solutions (such as Google’s AutoML and IBM’s AutoAI) which will automate and simplify the technicalities of data analytics (such as selecting the right input parameters to make a prediction) and of data management (such as managing data quality). These solutions will also allow users to use natural language to querying data instead of SQL.
FOUR: BUSINESS PROFESSIONALS WILL BE ABLE TO USE ANALYTICS THEMSELVES
Analytics & AI has traditionally been an area restricted to a few data scientists and data engineers. We already saw a few years ago the emergence of analytics translators, who are filling the gap between business subject matter experts and technical data professionals. Starting from 2020, more business professionals from various non-technical backgrounds will start to use big data in their own scope of work, thanks to new augmented and user-friendly solutions that don’t require technical skills.
FIVE: CLOUD DOMINANCE WILL INVOLVE THE DEMISE OF ANALYTICS OPEN SOURCE
Cloud computing has made forays for years already. In 2020, even the heavily-regulated banking and telecommunications industries will massively speed up their migration into the cloud, attracted by lower investment costs, simplified data management, and added-value tools that automate and augment Analytics. Companies will be captive of the cloud they choose and will ultimately leverage commercial platforms to manage their AI programs, signaling the beginning of the demise of open-source software, in the area of analytics.
SIX: AI TECHNOLOGY BUILDING BLOCKS WILL CONVERGE
Artificial Intelligence is a family of different building blocks used to make decisions (Advanced Analytics), communicate (Natural Language Processing), understand images (machine vision), and execute processes (Robotic Process Automation). Each building block requires very different specialized skills and tools, and companies typically combine these technologies to build use cases. From 2020, we will see the commercial maturity of frameworks that integrate multiple of these technologies. Even Blockchain, the Internet of Things (IoT), and Augmented / Virtual Reality (AR/VR) will be integrated with AI.
SEVEN: AI WILL PROGRESSIVELY BECOME LESS DATA-HUNGRY
AI techniques such as deep learning typically require massive data sets to be trained. Not all companies have these huge data sets and this is one of the handicaps for the development of AI in the enterprise. In 2020 we will see viable models that require less data for training by improving the learning capabilities of AI, reutilizing data among companies, or simulating new realistic data based on the available limited data.
EIGHT: NEW HUMAN-MACHINE INTERFACES WILL START TO BE MAINSTREAM IN AI
The human-machine interface for AI has not changed much in decades: computer screen, keyboard, and mouse. But this is changing fast. In 2020, we will start seeing mature commercial use cases using new human-machine interfaces, such as AR/VR (for example Augmented Reality in repair services on the field), smart glasses or lenses, microchips, holograms, or even brain-computer interfaces.
NINE: ANALYTICS WILL SIT IN THE C-SUITE
30 years ago Chief Marketing Officers were not in the C-Suite. 5 Years ago, Chief Digital Officers were also not part of the C-Suite either. Companies still hire hyper-technical analytics heads to start to lead their data efforts, instead of business transformation leaders with a mixture of technical and business capabilities. Therefore most AI programs fail because companies are unable to scale up pilots. This will start changing globally in 2020 when well-rounded Chief Data Analytics Officers start joining the C-Suite.
TEN: AI WILL BE INCREASINGLY A GEOPOLITICAL RACE
AI is becoming a ground for competition between countries. Some countries are affirming their data sovereignty by controlling data within their reach. Some are also intervening in the economy to direct the development of AI or even using AI in their interest to increase their power. Many countries have published national AI strategies: Canada, Japan, Singapore, and China, which intend to become the world’s leading AI power by 2030. Yet, two main players outperform all the others in the AI race: the US and China.
Do you agree or disagree? What are your views? Please comment below.
About Pedro URIA RECIO
Executive marketing professional with over 15 years of global experience in data-driven marketing, digital businesses, analytics, and AI, spanning technology, telecom, and financial services. P&L responsibility for quantitative marketing. Multifaceted, resourceful, and versatile former McKinsey consultant. Gifted in creating and coaching high-performing teams of marketing specialists, data scientists, and digital developers across multiple countries and cultures. Brave, creative and well-balanced between forward-looking marketing strategy and hands-on digital transformation. Chicago Booth MBA and certified machine-learning professional with a track record of bridging the gap between business objectives and technology. Committed technology author and international keynote speaker.
