Big Data and the AI Revolution
The 800-pound gorillas in the venture capital room are Big Data and AI, and the opportunities that will stem from them are many and diverse. Individually and in combination, these two technological developments are already having – and should continue to have – revolutionary impacts on our way of life.
Big Data already underpins several hugely successful business models, most notably Google and Facebook. It has found its way into a disparate array of industries, both independently and in conjunction with AI. Though popular concerns over personal data security must be considered when assessing any company leveraging Big Data, the range of information that companies can collect is likely to expand further as internet-enabled devices – referred to collectively as the “Internet of Things” – continue to proliferate. Meanwhile, the tools available to firms to interpret, analyze, and apply that data will continue to grow in sophistication.
We expect AI to similarly find its way into many different industries over time as its potential is better understood and harnessed. In fact, businesses are already moving rapidly to adopt AI systems.
Businesses moving quickly to integrate AI
80% – Share of companies at some stage of integrating AI
Estimated worldwide spending on cognitive and AI systems: 2019 - $37.5B, 2023 - $97.9B
$3.9T – Predicted business value to be created by AI by 2023
Sources: HFS Research and KPMG, Adoption Plans for Intelligent Automation (IA) Technologies in Organizations Worldwide, 2018. IDC, Worldwide Artificial Intelligence Spending Guide, September 2019. Gartner, Forecast: The Business Value of Artificial Intelligence, Worldwide, 2017-2025, April 2018.
In the shorter term, three of the most immediate and disruptive applications of AI we have identified involve new media, online education, and self-driving solutions.
Seven Types of Intelligent Automation
Most of the business applications of intelligent automation will draw from at least one, and usually more, of the following categories:
Artificial Intelligence – simulating human intelligence processes, including learning, reasoning, and self-correction, often underpinning other intelligent automation processes
Cognitive Virtual Assistants – automating information requests and customer service interactions
Computer Vision – interpreting, understanding, and responding to the visual world in real time
Machine Learning – interpreting and analyzing datasets to build mathematical models for making predictions or decisions
Natural Language Processing – processing, interpreting, and analyzing natural human language
Robotic Process Automation – automating and standardizing repeatable, rules-based business processes via software tools
Smart Analytics – intelligently analyzing and curating Big Data to provide business insights
Source: Categories from HFS Research, KPMG, Adoption Plans for Intelligent Automation (IA) Technologies in Organizations Worldwide, 2018.
Sample Disruptors*
TikTok Popular short video social app powered by AI China
DailyHunt AI-led multi-language news and content aggregator India
Pratilipi Self-publishing multi-language content platform India
DeepL AI-driven translation service Germany
Key Growth Sectors
New Media One of the most obvious and most fully developed applications of the mountains of data that companies can acquire from our digital activity is the creation of algorithms to tailor user experiences. This is true of multiple industries, but nowhere is the need more acute than in the twin realms of social and news media.
AI has already been applied in numerous ways on Facebook, including image recognition and the use of neural networks to tag photos – but to stay relevant in a highly competitive sector, social media companies will need to leverage their expanding data resources and continually improve their algorithms.
An excellent example of AI-led social media is TikTok, a short video-sharing social app from China’s ByteDance powered by an AI that curates content for the user.
For the news media, as the sheer volume of content continues to rise consumers will increasingly gravitate towards sophisticated but simple-to-use solutions to access it. Consumers have been empowered to select the news they consume via algorithm and aggregating tools. Two Indian start-ups that exemplify this trend are DailyHunt and Pratilipi, both of which deliver catered content in multiple Indian languages.
Yuan Fu Dao AI-powered online education platform for K12 students China
Byju’s Data-led company offering online video- and app-based personalized content India
OpenClassrooms Online vocational training provider France
Online Education Much of the criticism leveled at standardized education centers around its lack of flexibility given that every learner’s needs are unique. Online education apps offer companies the flexibility to shape education to a student’s needs by automatically identifying mistakes and adapting the materials presented. The student in turn benefits from a curriculum optimized to their level of knowledge and aptitude.
But much greater tailoring is still possible, down to the course-content level, along with techniques that can make learning more enjoyable.
Education is therefore an industry ripe for the application of AI solutions – and we expect significant growth in the number and sophistication of such educational tools. The sector has already grown significantly in recent years and is expected to expand at a 29% compound annual growth rate (CAGR) globally from 2017 to 2023 to hit $133 billion.³
China’s Yuan Fu Dao is one such leading start-up looking to exploit this growth. It uses an AI-powered algorithm to provide customized and adaptive learning programs for students across multiple subjects. In India, Byju’s model is similar, offering online video- and app-based personalized content for students, but the firm also harnesses app and web usage data to further enhance the user experience.
3 Market Reports World, Global Online Education Market -Trends & Forecast, 2017-2023, February 2018
Hesai Technology Advanced LiDAR provider China
Prophesee Real-time visual data processing system service France
Pony.ai Comprehensive solutions provider China
Intellifusion Algorithm developer China
*The deal summary, general partner, and/or companies shown above are intended for illustrative purposes only. While this may or may not be an actual investment or relationship in a HarbourVest portfolio, there is no guarantee it will be in a future portfolio.
Self-driving Solutions Beyond companies leveraging Big Data and AI to elevate and curate services, there are numerous deep tech companies that are tackling larger applications. One of the most impactful examples is creating solutions to support the development of autonomous vehicles.
Expected Growth of Self-driving Vehicle Market
Source: Allied Market Research, Autonomous Vehicle Market by Level of Automation (Level 3, Level 4, and Level 5) and Component (Hardware, Software, and Service) and Application (Civil, Robo Taxi, Self-driving Bus, Ride Share, Self-driving Truck, and Ride Hail) - Global Opportunity Analysis and Industry Forecast, 2019-2026, May 2018.
China plays host to many of the leading companies in this field, including Hesai Technology, Intellifusion, and Pony.ai. Hesai, for example, is one of the world’s most advanced providers of LiDAR, a laser emission method that calculates the distance to objects, effectively serving as the “eyes” of a self-driving vehicle.
The growth of mobility-as-a-service means car demand will progressively be driven by businesses that provide autonomous shared vehicles on demand – rather than consumers themselves. While widespread acceptance and usage of self-driving vehicles is still some way off, there is clearly a substantial potential market for providers, given the sheer scale of the automobile market.