Accelerating Automation

Accelerating Automation

Robert Whitby-Smith, Partner

The relentless pace of automation will increase in 2018. Whilst automation and technological innovation offer the best chance of solving the big societal issues including demographic change, access to healthcare, climate change and resource scarcity – we must be mindful of the social impact.

What is driving the accelerating pace of automation?

In 1958, the integrated chip was first developed. In 1965, Gordon Moore, the founder of Intel, predicted the computing power of an integrated circuit would double every year for 10 years. In 1975 he revised that to doubling every two years for the next 10 years. Extraordinarily these predictions have proven accurate. To put this level of exponential growth into context, if one invested £1 to buy a Moore’s Law VCT in 1958 and it doubled every two years, your single share would now would be worth over £1 billion. Some think this incredible growth cannot continue indefinitely, while others argue that quantum computing will soon enable computing power to accelerate faster. Whoever is right, computing processing power is now driving the main technique of machine learning – brute force big data analysis is enabling the automation of increasingly complex tasks. Meshing with this phenomenal growth in computing power include innovations in natural language processing using recurrent neural networks, augmented and virtual reality and additive manufacturing. These trends will lead to ever increasing digitisation.

How is automation impacting two traditional sectors – automotive and manufacturing?

We have a fascinating insight into the automotive sector through our recent investment into SBD Automotive. In 2007, parking assistance was introduced enabling a car to reverse park itself automatically at the press of a button. In 2010, Google put the first autonomous car on the road. Many doubted that autonomous cars would ever happen due to technological, cultural and regulatory barriers. Those barriers are rapidly falling. Consumer surveys suggest half of the public would now be willing to use an autonomous car. The US government has issued federal policy permitting automated cars on the roads. All the major OEMs have some variant of the four letter acronym CASE (Connected. Autonomous. Shared mobility. Electric), as their published strategies. In 2016, GM’s President stated that the automotive industry would change more in the next five years than the last fifty and in 2017 GM announced that it is “quarters not years” before it has its first autonomous car on the road. In 2019, Volvo will start supplying Uber with 25,000 “base cars” that include some form of autonomous driving technology on which Uber will overlay its own sensors and software to make the cars robo-taxis. Singapore is already trialling robo-taxis. Ok, so if autonomous cars are likely, what’s the next automotive revolution? Perhaps Atomico and Tencent answered that question in September 2017 when they invested $90m into Lilium Aviation to develop a five seat flying car.

The manufacturing sector meanwhile is seeing the impact of increasing automation with its use of robotics. Some factories are fully automated, with no human labour. Another innovation - additive manufacturing (3D printing) – is set to transform the sector further. GE and Rolls Royce are using 3D printing not just for prototyping but for producing titanium blades in the hot part of the engine. GE has announced its intention for all its aerospace components to be made using 3D printing in 20 years.

What challenges will these changes present for society?

The increasing digitisation of our lives presents many challenges, including the impact on employment, cybersecurity and privacy.

Employment - Machines are able to learn complex tasks quickly and are getting faster. In 1997, the world chess champion was defeated by an IBM supercomputer. In 2018, Google’s AlphaZero taught itself how to play chess from scratch in four hours before beating the leading chess computer. Recent developments in natural language processing and machine vision suggest machines are likely to match, and may soon surpass, humans at both image and speech recognition. This will enable algorithms to displace a high proportion of labour – with some estimates rising as high as 90%. There have been tech revolutions before, but nothing like this. In previous tech revolutions, machines have been tools. In the information revolution, the machines are becoming workers. In 2011, Marc Andreesen, a leading US VC, wrote in the Wall Street Journal that “software is eating the world” and “the spread of computers and the internet will put jobs in two categories: people who tell computers what to do, and people who are told by computers what to do”. Encouragingly, society is considering its response to this.

Cybersecurity issues are regularly front page news. Many readers will have been victims of phishing scams, identity theft, data loss or the temporary shutdown of their employer. However, the impact severity will increase as automation expands. Whilst having your online bank account raided can be costly, losing control of the acceleration, brakes or steering of your connected car could be fatal. As industries including defence, food logistics or nuclear power production become increasingly digitised, the risk of a rogue state (perhaps North Korea) causing a catastrophe increases. This is why cybersecurity will remain a key investment theme for Albion.

Privacy is an increasingly serious issue. Big data companies know what you buy, which websites you visit, which articles you read, your location, who your friends are and how you interact, how many hours you spend playing video games, which bills and taxes you pay (or not). Now advertisers know so much about you and your interests, they can be much more effective than the mass targeted broadsheets and TV. Facebook’s admission that the Russian government sent Pro-Trump targeted Facebook adverts to 126 million Americans during the last election campaign is disturbing.

Barring a popular revolt, we are entering an age where an individual’s actions will be judged by opaque algorithms and recorded in perpetuity. In 2017, China introduced the voluntary “Social Credit Scheme” which becomes compulsory from 2020. A citizen’s rating is calculated using an algorithm designed by Alibaba and the Chinese government to assess your qualities as a citizen. It is openly stated that what you and your friends say about the government, the economy and how you behave online will impact your eligibility for a job, a mortgage, the schools your children are accepted at. Perhaps this will include identifying “re-education” needs.

In summary, the direction of travel has been evident for some time. In 1964, US President Lyndon Johnson said “automation can be the ally of our prosperity if we will just look ahead, if we will understand what is to come and set our course wisely, after proper planning for the future”. Software businesses represent seven of the largest eight companies in the world by market capitalisation (Alibaba, Alphabet/Google, Amazon, Apple, Facebook, Microsoft and Tencent) and the pace of automation is accelerating. There has never been a more exciting time to be investing in automation and innovation, but we need to be mindful of the social impact.