The AI Paradigm Shift
AI and its dimensions
We are living in a period of rapid change, where AI will transform every aspect of our lives and the fabric of our society. It will affect most human activities from supply chains to healthcare to education, manufacturing, entertainment and even space exploration.
It is key to understand the dimensions of what AI really means. This “new” technology has been around for 6 decades, starting with rule-based systems and currently evolving to machine learning and deep reinforced learning (machines training machines).
These technologies are able to enhance human capabilities, including natural language systems, and the myriad of applications that have taken over our devices and are getting substantially better every year as data becomes more abundant and freely or easily accessible for early-stage companies and innovators.
Source: Thompson Reuters
This is not just a popular topic in the investment ecosystem. At the World Economic Forum 2018, in Davos, AI was one of the two major discussion topics, with clear ramification across industry sectors, with a potential for disruption at national, regional and global level.
There is an associated fear, and growing body of literature and research trying to quantify the side effects of AI and automation more broadly, as those technological advancements reach their respective tipping points.
Among some of the findings of such studies are systemic unemployment, widening income gap, optimal consumption and an attack on the open competitive market. These headlines lead many experts in our space to imagine very different realities for the next generations. From technological socialism to a singularity where we all enjoy the endless potential of General AI, all the way to a dystopic landscape of radical selection, a rise of robots, a contraction of the global population and world where being human will not be a quality.
It is uncertain which of these scenarios or a mix of potential realities will unfold of the next three decades, what it is clearer than before is that AI technology is now emulating human characteristics and qualities, such as creativity, vision, speech, analytical abilities with multiple streams of data (structured or not). Further, it is moving closer to simulate emotions such as empathy.
Why is this happening now?
The technologies have been around for a while, but why are we getting excited about it now. There are broadly 5 trends that are converging, creating a nurturing platform for experimentation and innovation:
The cost of computational power (Moore’s law), had a positive effect of lowering the entry barriers for entrepreneurs and researchers, leading to an acceleration of innovation across the globe. The economics of bytes interacting with the world of atoms. Cloud computing, lowering costs of data storage, the adoption of smart devices and the exponential increase in data generated, made this an even better momentum.
Evolution Data technologies and data systems. Further enhanced by established statistical analysis, an evolution from probability to Markov simulations pattern recognition, and decision systems as a subset.
The Future of AI
It is expected that many such AI technologies will start to be applied across sectors, commercialised, and later upgraded. A process fuelled by the exponential increase of investment across AI, culminating with 2017, when over $11bn was invested across 1174 deals, doubling from 2016 in absolute and relative terms (data from Pitchbook AI Report 2018).
This sector is perhaps the hottest in terms of exits and acquisitions, with over 133 such deals in 2017 and over $18bn invested. It seems that both corporates and other sector players are starting to consolidate across verticals.
Without our fund we already have over 25 such technologies, in 2017 we have been ranked 4 in our top for making 12 investments in the AI space alongside our co-investment partners, including Albion Capital, Downing Ventures, and Seedcamp.
We are witnessing in real time the changes that could potentially rethink every industry and area of human activity.
Among some the technologies expected to emerge, enter the market and peak, over the next three decades are:
- Artificial General Intelligence
- Cognitive Computing
- Conversational user interface
- Autonomous vehicles
- Virtual Reality
- Robotics and Smart Dust
- Emotion reactive AI
Source: Gartner Hyper Cycle for Emerging Technologies 2017
It is not the Luddite approach that will save our roles in our society but the ability to adapt to these changes, to seize the moment, and ride the sixth Kondratieff wave.
This is a great opportunity to create the future like no other generation before we ever could, to define the new role of the state and of corporations, to contemplate social impact in every area of activity, and to utilise our greatest resources to ensure future generations will have something to inherit.