This post is part of series that focuses on the disruptive potential, commercialisation and adoption of the key new technologies underpinning the Fourth Industrial Revolution. The first focused on the Internet of Things and the new smart connected products, and coming posts will consider the concurrent advances in adjacent technologies including Augmented Reality, 3D Printing and Blockchain which are amplifying the scale of the transformative impact and the merging of our real and digital worlds.
What is Artificial Intelligence?
Artificial Intelligence (AI) is an umbrella term for a spectrum of capability endowed by a number of different technologies to both physical machines or 'robots' and to soft systems. From robotic process automation (RPA) to natural language processing, machine learning and deep learning, AI technologies variously allow machines to sense, comprehend, interact and learn from the data captured from every smart connected product and device connected to the internet.
Disruptive Potential of AI
The popular media focuses on a dystopian vision of widespread workforce automation, and the disruptive potential of AI is perhaps unprecedented. Undoubtedly many roles can be fully automated, but there will be a seismic shift towards collaborating with machines and ‘cobots’ that augment our human capability perceptually, cognitively and physically, transforming ways of working and living.
New product offerings enabled by smart connected products and devices, to which machine learning can be applied in real time in the cloud, will increasingly interpret context to proactively and predictively deliver what we individually value at any given time.
Commercialisation of AI
Expertise in non-learning technologies, such as robotic process automation, is well advanced. For machine learning technologies, the level of commercialisation is unevenly distributed. The technology platform giants have enjoyed significant advantage from the enormous digitised consumer data sets captured over the last decade with which to build models. Their offerings dominate the current landscape within which longer established and less visionary enterprises struggle to wrangle and make sense of data from disparate legacy systems, but synthetic data is providing opportunities to close the gap.
Furthermore, the data scientists to develop models and train AI algorithms are in short supply and concentrated within the technology giants or promising start-ups that can reward their talent. Businesses without the in-house technology expertise are forced to seek external help to keep pace and maintain customer relevance.
Market Adoption of AI
AI has delivered and driven our ever-increasing consumer expectations of intuitive, personalised products and friction-less experiences. We readily embraced Amazon’s recommendation systems, Spotify’s playlists, Google’s navigational directions and, increasingly, virtual assistance from Siri and Alexa. Only now are we experiencing unease at the reach of this new capability, and the loss of control over our personal data.
For organisations, the lack of data and expertise can be compounded by low levels of AI literacy at the executive level, whilst inexplicability of the factors and reasoning behind the ‘black box’ of algorithm-generated recommendations can leave businesses exposed to regulatory or ethical backlash. The critical cross-disciplinary skills to embed ethics, counteract bias, and hold humans from harm from AI are only now emerging in response to identified failings.