The increasing digitalisation of industry is bringing together advances in sensors, through the Internet of Things, other immersive technologies, Big Data analytical tools and machine learning/artificial intelligence. Together we regard these as part of the “4th Industrial Revolution”.
At the simplest level, the pandemic has accelerated several digital technology trends, leading to more working from home and increased use of online collaboration tools such as Microsoft Teams and Zoom. A notable development has been the increase in virtual GP consultations, which can be expected to continue after the pandemic has passed (interestingly, text and telephone have been as useful as video). Online learning – and “blended” learning – is becoming the new norm.
The “Internet of Things”, networks of connected devices, is dominated by applications in manufacturing (44%), followed by asset management, smart homes and freight management, with many innovative applications in health care, retail, energy, and agriculture. IoT sensors enable:
- remote control, predictive maintenance and energy management;
- “wearables” for health monitoring, in hospital or in the community – eg smart contact lenses can detect glucose levels and deliver insulin as necessary;
- precision farming, determining exact levels of water etc to be administered;
- smart cities, managing congestion and pollution.
Forecasts of the number of connected devices run into the tens of billions, even trillions, but most development effort is currently focused at the network level. Although many future applications are touted, the reality may be more mundane, with smart homes and smart cities being exaggerations of the actual benefits delivered. Intelligent chocolate dispensing machines aren’t a major societal development.
Challenges still facing IoT include latency, integration and especially cyber-security of the communications networks themselves, and user acceptance issues around privacy. Suggestions that the insurance industry could use sensors to help set premiums are experiencing pushback – Belgium has introduced a law banning insurers from using health monitor data.
IoT generates huge amounts of data, so has a natural link to “Big Data” analytics. A now-mainstream definition of big data is the three V’s:
- Volume: from business transactions, smart (IoT) devices, industrial equipment, videos, social media and more.
- Velocity: the need to deal with very high volumes of data in near-real time.
- Variety: structured, numeric data; unstructured text documents, emails, videos, audios, etc.
But the benefits of Big Data come from analytical tools which enable users to identify statistical inferences and correlations. Traditional analytic techniques struggle to handle such vast amounts of data, so new parallel processing tools and techniques
such as Topological Data Analysis are required to explore the “shape” of the data.
This brings us on to Machine Learning and Artificial Intelligence. Sectors where AI is particularly beneficial include:
- Law: many areas of law require relatively straightforward analysis of a large number of documents – planning applications, personal injury fraud;
- Medicine: similarly AI has found many applications in medicine, in cancer detection, drug development, predictive diagnosis and triaging, as well as more general clinical decision support systems . The pandemic has stimulated many developments in this field, however a recent review suggested that few systems are currently matureenough yet to show operational impact.
- Retail: covering a range of areas such as sales and CRM applications, customer recommendations, logistics and delivery and Payment Services
AI is penetrating journalism – though sometimes with unfortunate mistakes, publishing a photo of the wrong Little Mix singer. Thomas Frey (one of the authors of the report on AI and jobs) identifies areas of the arts and entertainment that AI will transform. AI creating “deep fakes” – manipulated or entirely manufactured images and recordings of people that are very hard to detect – will create a further challenge in an era of “fake news”.
A related application is facial recognition, though its use in policing has been challenged because of interference with human rights and its inaccuracy, especially with ethnic minority faces. Russia is developing a “Minority Report”-style system for identifying people behaving aggressively: “pre-crimes”. The Chinese government is accused of using facial recognition to commit atrocities against Uyghur Muslims.
These last examples indicate the ethical issues associated with AI. AI-driven weaponry is a major area of concern. And if China is investing more in AI than any other country, then its ethical principles will implicitly become the dominant ones.
AI also underpins developments in Automated Vehicles: driverless cars, robotaxis. Again, despite ambitious claims from Tesla and Baidu, reality may more challenging as regulators take a hard line on accidents. Some suggest that driverless would still have as many as two-thirds as many crashes, and “safety drivers” in such cars may still be prosecuted.
At the top end of advances in the field of AI robotics is Popper, a semi-humanoid robot designed with the ability to read emotions based on detection and analysis of facial expressions and voice tones. The pandemic stimulated the use of telemedical robots as a way of reducing physical contact.
As we observed in our blogpost on geo-political trends, there are a range of views on what AI does to employment. Susskind and Susskind in their book The Future of the Professions: How Technology Will Transform the Work of Human Experts, predict the decline of today’s professions (doctors, teachers, accountants, architects, the clergy, consultants, lawyers, and many others), not just clerical jobs. Others argue the reduction in routine tasks is to be welcomed, leaving roles requiring empathy and relationship-building to humans – but AI is moving into these areas too with mood detection.
In the area of technological advances, it is important to maintain a degree of scepticism. The Gartner Hype Cycleillustrates how often, after a burst of enthusiastic promotion, innovations can settle down to a specialised but lucrative niche – the “plateau of productivity”. Nonetheless, the more dramatic claims can be a useful way of shocking people out of their current comfort zones and encouraging them to explore alternative scenarios.
Written by Huw Williams, SAMI Principal
The views expressed are those of the author(s) and not necessarily of SAMI Consulting.
SAMI Consulting was founded in 1989 by Shell and St Andrews University. They have undertaken scenario planning projects for a wide range of UK and international organisations. Their core skill is providing the link between futures research and strategy.
If you enjoyed this blog from SAMI Consulting, the home of scenario planning, please sign up for our monthly newsletter at email@example.com and/or browse our website at https://www.samiconsulting.co.uk
Image by 3D Animation Production Company from Pixabay