Technology has evolved at an incredible rate over the last 30 years. My smartphone has over 488 thousand times as much memory as the first mainframe that I booted up in 1981. Processing power and storage is cheap and the networks that transfer our data are fast enough for us to worry no longer about sending large files across the internet. Over the same time period, however, the human being has changed very little. We continue to get very excited about technology that is going to ‘change our life’, and, as a result of that, we rush to implement solutions without thinking of all the possible consequences. We are as shocked and surprised when technology fails us now as our ancestors did when the ‘unsinkable ship’, the Titanic, sank in 1912.

We also have a worrying inability to remember the lessons learnt from disasters that resulted from corners cut or poor decision-making. We let deadlines override the common sense that tells us to hold back on implementation and deployment. Nobel Laureate Richard Feynman was a member of the Rogers Commission team, which investigated the 1986 Challenger shuttle disaster. The night before the disaster, engineers had advised against launching, as the O-rings were known to fail at low temperatures and the forecast temperature for the following day was –2C. However, timing was critical to enable the first teacher in space to give a lesson during school hours, and the fear of reputation loss from a delayed launch was front of mind for the decision-makers. Richard Feynman stated, ‘For a successful technology, reality must take precedence over public relations, for nature cannot be fooled.’¹ The Challenger disaster was 30 years ago, and yet we continue to see technology failures triggered by poor decision-making, resulting in loss of life.

The ease of sharing data and analysing joined data sets that has resulted in the Internet of Things (IoT) – a rush to embed microchips in everyday objects and to set those everyday objects up to share data with each other – represents a huge leap forward in the potential of technology to provide innovative connected solutions. As for all great leaps that have gone before, the IoT presents opportunities to deliver great value services but also exposes great risks. There are some fantastic applications – solutions that enable elderly people to live longer in their own homes, sensors that combine to provide early detection for fires and floods, systems that supply accurate real-time information for buses and trains, and so on. Town planners and technologists dream of smart cities where everything is interconnected and huge savings in power costs, as well as fewer traffic jams and burglaries can be achieved. And for some cities, this dream is becoming a reality.

RISKS ASSOCIATED WITH THE INTERNET OF THINGS

With the aim to deliver great value in mind, let us consider some of the great risks and how they can be mitigated. First, we need to understand that, where in the past a system failure would have affected the users of that system only, we now need to consider the users of interconnected systems. If my ‘intelligent’ light bulbs were to communicate with each other in open text across my wireless network, and they were to be hacked, it is almost certainly not just my lights that would experience a black-out. We need to build security solutions into our IoT, where one weak link can expose multiple systems and devices and put many at risk.

Another great risk is our over-confidence in new technology and our inability to learn from previous mistakes. Let us consider the driverless car. In under two weeks, there were two related news stories. The first story told of a recall of cars by one particular manufacturer due to ‘faulty software that can briefly shut down the engine’.² It should be noted that ‘both the engine and electric system start up again immediately after shutting down’, but such an error would be momentarily frightening. The second story, concerning the same manufacturer, spoke enthusiastically of driverless cars being available on public roads within a year.³ We have built the notion of a driverless car into an infallible product that will save us time and money and reduce pollution – and it may well deliver on all those points. However, if we neglect to plan for software and hardware errors and ignore single points of failure, we will be releasing fleets of ‘Titanics’ onto our roads.

We need to build our IoT with robustness, quality and business continuity in mind. We will have an increasing requirement for professional testers who understand the interconnectedness of the systems they are testing and the consequences of the results they see. And we need to listen to and act on the advice from our designers, developers and engineers who can see the potential weak points in our connected systems.

The final ‘great risk’ that I would like to consider concerns the ethical or moral use of IoT solutions, where the potential for value and risk is determined by the application of the solution and not the solution itself. There are applications of the IoT that could be used for good or evil. Drones can be controlled remotely and used to deliver pizzas or identify maintenance areas in a nuclear power station where humans could not enter safely. Yet drones have also been used to deliver drugs over prison fences. Full marks to the ingenuity of the Dutch police, who train eagles to bring down ‘enemy drones’.⁴

I would like to think that there was a simpler way of addressing the problem of ethical use of IT systems before release and deployment. We need to build our IoT solutions with a consideration of the consequences of how our combined technology could be used and, as far as we are able to, protect against harmful use.

So, in conclusion, let us learn from the errors we have made over the years in rolling out single new technologies before we roll out multiple connected technologies in the form of IoT solutions. Let us look for value and innovative solutions, but let us also consider mitigating against potential negative uses and testing solutions thoroughly before we deploy.

REFERENCES

1. Feynman, R. P. ‘Appendix F – Personal observations on the reliability of the Shuttle’ Available from: https://science.ksc.nasa.gov/shuttle/missions/51-l/docs/rogers-commission/Appendix-F.txt [27 March 2017].

2. Press Association (2016) ‘Volvo recalls 59,000 cars over software glitch’. The Guardian. Available from: www.theguardian.com/business/2016/feb/20/volvo-recalls-59000-cars-over-software-glitch [27 March 2017].

3. Volvo Car UK (2016) ‘Volvo Cars to launch UK’s largest and most ambitious autonomous driving trial’. Volvo Car UK Newsroom. Available from: https://www.media.volvocars.com/uk/en-gb/media/pressreleases/189969/volvo-cars-to-launch-uks-largest-and-most-ambitious-autonomous-driving-trial [27 March 2017].

4. Agence France Presse in Ossendrecht (2016) ‘Eagles v drones: Dutch police to take on rogue aircraft with flying squad’. The Guardian. Available from: https://www.theguardian.com/world/2016/sep/12/eagles-v-drones-dutch-police-take-on-rogue-aircraft-flying-squad [27 March 2017].