Skip to the content

Tech Trends 2018: The year in innovation
personChris Middleton eventJan 2, 2018

Tech Trends 2018: The year in innovation

Hack & Craft News present a rundown of eight technology hotspots, strategic topics, and things to watch out for this year.

1. GDPR:
Consent as a currency... or the new millennium bug?

For most organisations, 2018’s big strategic tech challenge arrives in the Spring. The General Data Protection Regulation (GDPR) comes into force in May, fulfilling a dream of the Web’s prime mover, Sir Tim Berners-Lee: citizens taking back control of their data on their terms. Consumer-sponsored CSR, perhaps.

The jury is still out as to whether GDPR will genuinely transform data privacy, transparency and security, or whether – as some have claimed – little will change in the real world. In a recent interview, SugarCRM CEO Larry Augustin said he believed some organisations will simply ignore the regulations and call governments’ bluff.

“There’s a set of companies that are saying, ‘We don’t know what it means to be compliant, so we’re just not going to do it.’ They’re going to take an aggressive interpretation and say, ‘Look, this is ill-defined, it doesn’t apply to us that much’. But that doesn’t work with GDPR. They’ll have to comply eventually, but they’re willing to take that path.”

Whatever the immediate effects of GDPR may be, it makes critical changes to how organisations must treat personal data, by putting informed consent and transparency at the heart of the regulations, alongside privacy by design and the right to have data permanently erased.

So forward-looking organisations shouldn’t see the regulations as an obstacle. In the future, consent will be a de facto currency, backed by data ‘gold’. This alone should encourage enterprises to unlock innovation, not stifle it or moan about bureaucracy.

The blockchain/crypto consideration

Blockchain and cryptocurrencies are two further hotspots for 2018, and similar technology models can be applied to consent: for example, consider consent as a blockchain-enabled means of transacting business, backed by deep reserves of data gold. 

For example, US startup SimplyVitalHealth is using blockchain to put patients in charge of their own medical data, allowing them to sell it to research institutes, while matchmaking them with private care providers.

A technology engine for managing individuals’ consent in any digital relationship is, arguably, long overdue, perhaps based on an alternative concept: personal APIs – like a digital rights management platform for citizens’ Ts and Cs. Such a system might attach conditions to individuals’ consent and embed them in the API.

For example, Citizen 1 might say, ‘You may use this personally identifiable data [PID] for this purpose, and for these causes that I support, but not for X, Y, or Z.’ Meanwhile, Citizen 2 might happily give his PID away in exchange for programmatic advertising, but at least it would be an informed choice.

Companies interfacing with each citizen’s key, via our notional consent platform, would be entering a legal obligation to honour those terms, finally giving something back to the consumers that some organisations have regarded as fair game. Meanwhile, companies would be rewarded for good behaviour, by increasing their stocks of data ‘gold’.

Whatever the long-term technology solution may be for managing consent, the underlying question is simple: if you challenge or ignore GDPR, what are you trying to hide, and what message does this send to your customers?

In a long game, the companies that champion consent will win.

2. Customer service automation:
Automatic for the people

Offshore call centres and interactive voice recognition have long been irritants to many, but they are fast being replaced by chatbots, AI, and services that link robots to IBM’s Watson in the cloud, for example, enabling people to have natural language conversations with machines, backed by industry-specific datasets.

But that’s not to say that AI systems and chatbots should be strategically adopted as people replacements. Most vendors in the cognitive services space see the technologies in terms of augmentation, not replacement: man plus machine, not vs machine. Vendors such as IBM, Microsoft, and say their AI tools are designed to help people make better decisions, while deepening human relationships.

Does sweeping aside costs and people make the customer experience better? That’s a business decision, not a technology one. Strategists should put the customer first, then bring the right supportive technologies onboard. Automate services that add no value, and optimise the human experience everywhere else.

Meanwhile, Alexa, Siri, and other digital assistants are becoming familiar voices in the home. What lies behind them is a more important question than most people realise. In the case of Amazon, for example, its retail, fulfilment, logistics, and distribution operation links with its Web Services division, via an innocuous looking speaker in your living room.

In such an environment, trust and transparency will be essential: what goods is your home hub ordering, from whom, and why? One thing is certain: antitrust regulators will be watching, and ­ again – the future belongs to those who see consent and openness as a currency.

Source: Adobe Stock

3. Ethics man:
Good behaviour as a differentiator

As we’ve already seen, consent, trust, and transparency will be the watchwords of the fourth industrial age. Nowhere is this more apparent than in the worlds of AI, machine learning, and robotics, where serious concerns exist about the risk of machines automating bias, discrimination, and other human problems, as this report explains. 

To overcome these challenges, some have suggested that ethical software and systems development should be adopted as a matter of principle. For example, in 2017, the RSA and YouGov published a wide-ranging report, The Age of Automation, which suggested that coders should sign the technology equivalent of a hippocratic oath.

Another solution is to increase diversity in the technology sector. Only 17 per cent of people in STEM careers are women, and most developers are young, white males. Sometimes the systems that are developed in these environments reflect the teams’ own lack of diversity.

4. Distributed logistics and manufacturing:
Meet your new PAL

For years, Western businesses have relied on mass low-cost overseas manufacturing, offshore outsourcing (offshoring), and global distribution to keep costs down and profits up. However, these macro/monolithic processes rack up thousands of miles and millions of tons of carbon to get cheap goods to the West – at glacially slow speed.

The same processes take place within nations, as companies struggle to compete with the next-day delivery promise of highly automated retail and fulfilment organisations, such as Amazon: massive centralised warehouses, and fleets of trucks shipping goods from one end of the country to the other.

However, monolithic manufacturing and distribution could be rendered obsolete in the medium to long term, replaced by on-demand manufacturing and a grid of modular, micro supply chains that are more personalised, automated, and localised – ‘PAL’, an acronym coined by transformation consultant Sean Culey.

A broad range of technologies are emerging and converging, and these are automating not only labour/manual and industrial processes, but also mental ones.

These technologies include:

  • AI and machine learning, which are being built into countless enterprise apps and platforms, including Google. IBM has reoriented its entire business around cognitive services, and Microsoft is moving in the same direction.
  • Big data analytics.
  • Small, programmable industrial robots, such as those provided by British startup Automata. Gone are the days of dumb, single-use machines. With smaller, cheaper, faster robots hardware emerging in the years ahead, the apps that run on them will be the key.
  • ‘Cobots’ that work alongside human workers – like Ocado’s trial of ARMAR-6 humanoid robots in its warehouses, as part of its SecondHands project. More details here
  • Mass automation of warehouse processes, as spearheaded by Amazon and others.
  • 3D printing, aka manufacturing in people’s homes. In this context, IP and schematics are the real products: the idea of something, rather than the thing itself (see Digital Twins, below).
  • Drones, autonomous vehicles, and driverless delivery fleets, such as those being trialled by Tesco, Hermes, and others. Even Uber’s long-term future will be driverless, judging by its multimillion-dollar investment in the technology.

All of these technologies will combine to manufacture goods on demand and deliver them swiftly to your door.

And PAL doesn’t stop there: the sharing economy is transforming logistics in other ways. A number of companies, including Flexe in the US and Stowga in the UK, are applying the Airbnb model to warehousing, opening up (inter)national networks of spare warehouse capacity on demand.

Run towards the future

So how does all of this add up? Just as AI- and automation-enhanced aerial farming is bringing food production closer to the mouths that need feeding – another 2018 tech hotspot – so PAL manufacturing and logistics could do the same for all types of goods.

Take the hyper-competitive world of sports shoes as an example. Company X may have millions of identical trainers manufactured in Asia, then ship them worldwide: a process that takes weeks or months, at a colossal environmental cost. Whereas Company Z might custom-make a single pair of shoes for you on demand and deliver it box-fresh to your doorstep the next morning, slashing carbon emissions and incurring no extra costs. Which option do you choose?

Adidas certainly understands the concept, with its Futurecraft line of sneakers. The company’s new automated, robot-staffed Speedfactories are opening throughout the world, allowing it to fine-tune products to local tastes, while reducing time to market and to doorstep. Glimpse the future of PAL manufacturing and distribution here.

Ultimately, the strategic goal of ventures like these becomes the app in the customer’s pocket, enabled by customer profiles, big data analytics, and ERP at the supplier’s end. Every item is custom-made to your preferences, and delivered automatically or autonomously the next day from a local facility, not from the other side of the world.

In this way, it is just as economical to make and deliver one item as it is to manufacture a million. Waste is massively reduced, as is the carbon impact, and the customer gets exactly what they want within 24 hours. Everyone wins.

It’s all about the data

Sensors and the Internet of Things (IoT) are part of this trend too. For example, British rapid-prototyping company RPD International has put sensors in everyday household items, allowing it to understand how people in, say, the Philippines might use a broom or a chair differently to someone in the UK or Brazil. Those products can then be custom-modified for local markets.

In the UK, where long-term trading relationships are up for grabs, PAL manufacturing and distribution could reinvigorate long-depressed parts of the country.

And the employment picture is not as simple as ‘one robot in, 15 people out’: Amazon has created nearly 100,000 new human jobs in its recent journey towards automated fulfilment, which contradicts the mass-unemployment scenarios presented by some newspapers.

So what’s the key to this radically different future? Be bold and imaginative, and above all have the courage to rethink your business from the ground up. If you don’t, someone else will.

Source: Adobe Stock

5. Open innovation:
We’re all in this together

The days of ‘Mac, PC, or Linux?’ are now so long gone as to seem ridiculous. In 2018, innovation after innovation will teach us that the service is the real product, and not the hardware that delivers it in our pockets, homes, offices, or factories.

As we’ve seen, even industrial robots are becoming smaller, faster, and more programmable. So much so that the hardware will rapidly become commoditised as the app store model emerges. As happened with mobile phones and tablets, the robotics sector will eventually be about what people can do with these services rather than the machines themselves.

For service providers, open development and technology agnosticism will be the drivers of innovation: it’s all about the service, the standard communication protocols, and the shared values, not the technology.

The trend manifests itself in countless other ways, too, such as collaborative networks.

For example: OpenAI is a non-profit research group beating a collaborative path towards safe artificial intelligence; the Open Data Institute, co-founded by Sir Tim Berners-Lee, looks at how open data sets can be shared and exploited for the design of smarter communities; and DesignSpark is an online community offering free tools, resources, and support for engineers. A sharing economy in which collaboration and knowledge exchange are encouraged, not squatting in data silos.

There are other benefits to this mega-trend. If the IoT is a giant robot, in effect – eyes and ears in every home and office, and hands in every factory – and PAL supply chains reach critical mass, then everyone could access these facilities in the cloud. Rapid prototyping, testing new ideas, launching new businesses, and so on, could all take place within open, collaborative communities and with technology-agnostic development.

6. Digital twins:
A blueprint for everything

‘USS Callister’, the first episode of Series 4 of Black Mirror (Netflix), starts out as a Star Trek pastiche, but swerves into much darker territory. At the heart of this dystopian tale is the concept of digital twins. Or rather, a futuristic spin on the idea, in which digital copies of a company’s employees, created from their own DNA, are trapped in a computer simulation. A commentary, perhaps, on those tech visionaries who have suggested we may be living in a giant computer programme.

At present, digital twins are rather more mundane: 3D digital representations of physical objects, which can be exploded and examined component by component – a logical output of the 3D/CAD design processes that created them. They serve a critical purpose, especially when combined with other technologies, such as AI, analytics, enterprise resource planning (ERP) and enterprise asset management (EAM).

Perhaps the most advanced example can be found at CERN, where the Large Hadron Collider (LHC) remains the largest machine in human history. Every one of the 100 million components that exist in two million pieces of equipment on CERN’s campus has a digital twin stored on file, from the smallest screw, bolt, or widget, to the largest sensor, particle detector, or supra-cooled magnet. Each twin can be called up onscreen, analysed in minute detail, and – in virtual form – have components taken out, examined, upgraded, or replaced.

Linked with AI, analytics, ERP, and EAM, digital twins mean that engineers can predict or trace faults at component level, plan obsolescence or repairs in advance, and get replacements manufactured to order.

So think of digital twins as enabling a 3D library of data about everything in the physical world – a library that links with 3D printing, too. Digital twins can tell you what’s wrong with real objects, teach you how to fix them, and provide schematics for replacement parts.

Consider other applications, such as the component parts of vehicles, office blocks, oil rigs, power stations, factories, hospitals, transport networks, or cities, and the use cases for combining digital twins with AI, analytics, ERP, EAM, sensors, and smart supply chains should be clear.

So might digital twins one day start believing they’re real? It’s conceivable: DNA is just another data storage medium, after all, and understanding the human brain might not be necessary if it can simply be scanned and replicated. Crossing the great divide between the real world and the virtual one can’t be too far off in the future.

Might we all have digital twins, so that replacement parts can be 3D printed when we succumb to illness or accident? We’re already several steps along in that journey.

Source: Adobe Stock

7. Infrastructure security:
Living on the edge

If we learned one thing in 2017, it was that malicious agents have not only hacked IT systems, but also our belief systems, via troll farms that spread dissent on social platforms and sought to influence elections and referenda.

As the IoT spreads and becomes the Internet of Everything, security takes on entirely new dimensions.

While it’s tempting to assume that all our future processing, intelligence, applications, and so on, will somehow reside ‘in the cloud’ – that fog of US marketing that describes land-based data centres – the reality is that cloud platforms are just one element of a communications infrastructure that includes hardware, software, edge computing, and a distributed core.

The pendulum has been swinging back towards distributed systems, and away from centralised cloud services, for some time, and the edge environment is where much real-time AI and IoT processing will take place. The reason is that, with an estimated 30 billion connected devices online by 2020 alone, a mass of-in-memory processing will be essential, with other data-crunching carried out near the source.

This is because of poor network speeds and connectivity in some locations, and cloud’s latency in time-critical applications, such as data triage, real-time analytics, machine-to-machine instant messaging, a smart vehicle’s need to avoid a collision, or augmented/mixed reality’s requirement to overlay data on top of the real world in real time.

Speaking at an IoT event in New York last October, Dell CEO Michael Dell said, “With the cost of a connected node approaching zero dollars, the number of them is exploding. We’ll soon have 100 billion connected devices, and then a trillion, and we will be awash with rich data.”

But we’ll also be awash with new types of security threat. As countless vendors rush to capitalise on the IoT, the attack surface is exploding: a Big Bang of potential targets.

With applications and operating systems occupying most enterprise security professionals’ time and attention, hardware and firmware are increasingly in hackers’ sights – especially with OEM hardware that uses common components from outsourced manufacturers. Trust increasingly needs to be assured at metal level.

And then there are those billions of IoT devices.

A few years ago, IBM white-hat researchers demonstrated that it was possible to access an office building’s data centre via an insecure smart lightbulb, and to disable a driverless car’s brakes using a hacked MP3 file. Since then, scare stories have spread about Facebook, Amazon, Google, et al, spying on users via their mobiles, smart TVs, and home hubs. Some of this may be post-millennial angst about the speed of technology change, but these types of threat are real.

If a robot can be described as the eyes, ears, and hands of the internet, then those sensors and limbs can also be accessed remotely, placing a spy in every home. Edge intelligence will be the key to battling this problem, and companies such as Zingbox are looking at different aspects of security and trust in the IoT, with a focus on detecting problems before they become systemic failures.

Zingbox CTO May Wang explains how the IoT will be like the human body: “We all know a heart attack can be a catastrophe, but it can take days, even years, to build up. But if we can monitor and detect early signs of an attack, we can prevent it.”

Monitoring network health

Healthcare is just one of many sectors in the vanguard of this new security challenge, as we’ve seen from ransomware attacks on the NHS. But that was a problem of obsolete systems, which are common in the medical world.

According to Wang, recent security breaches in the US have been of a much more troubling and insidious nature. They include:

  • hospital environmental systems being controlled remotely by hackers
  • hacks of CT scanners, and systems that control patients’ drug dosage
  • health records being stolen via an unsecured x-ray machine, and more.

Just as an anonymous troll might victimise someone on Twitter, so an anonymous hacker might one day kill someone remotely – if he hasn’t done so already.

A new type of security is needed in such a world, something that can’t always be built or retrofitted into the devices themselves, especially in sectors that are heavily regulated, or which use outmoded equipment.

AI, machine learning, monitoring, and detection, together with automatic discovery and identification, will be the only realistic approach to IoT security in the long term: systems that detect unusual profiles and infer unusual behaviour as it emerges.

We may be days or months away from the first major IoT security breach – who can tell? Will it take place in a home, a school, a factory, a transport network, a nuclear power station, or a hospital? That’s impossible to predict, but systems need to be in place to identify unusual behaviour as soon as it happens, whenever the hardware and the code that runs on it can’t be trusted.

Source: Adobe Stock

8. Corporate innovation models:

We’ve seen some of the ways in which 2018 will be marked by new technology and strategic business challenges. But how can organisations themselves innovate?

In recent years, we’ve come to believe that innovation comes from superstar CEOs – the Steve Jobs or Elon Musk model of outlier charisma, vision, and drive. But no amount of vision will succeed if the organisation isn’t structured to deliver it, or if the vision is rooted in a world that’s vanishing.

In this report we’ve already explored how some centuries-old business models may be replaced by modular networks of local functions; even banking may be undermined by cryptocurrencies. Meanwhile, governments fret about encryption, even as blockchains and secure communications spread. The key is to step back and see what’s really happening, and the signs can be found within our own organisations.

Most of us are aware that the idea of ‘the office’ is fast disappearing. Flexible and remote working, mobility, and more, have turned many organisations into sets of ideas that are shared by flexible teams, and cloud-based collaborative tools support this approach. Put another way, digital technologies tend not to support centralised, top-down, hierarchical structures, even though many such organisations continue to benefit from them.

Meanwhile, a failure of policy in some governments has been to try to shoehorn 21st Century technology into 19th Century industrial models, to maximise the benefits for fewer and fewer people – a reductive focus on productivity stats rather than on how and why people are productive.

One example of this appeared in 2017, when right-wing think tank Reform (supported by the Prime Minister, Jeremy Hunt, Tony Blair, and others) published a report on AI and automation in the UK public sector. The document praised the technologies’ ability to slash 250,000 jobs from government and force workers to compete via reverse auction (bid to work for less money) in the gig economy. Concrete proof that ‘for the few, not the many’ is an active ideology, and not a left-wing scare story.

Yet many of the most innovative organisations today marshal the many, so that the many can benefit. Digital tools naturally support a sharing economy, along with localised, personalised, automated systems, matching goods and services to local needs and preferences. Heat-seeking teams are thriving, linked by collaborative platforms that focus on iteration towards market fit rather than ‘the big launch’.

So might even the concept of leaders, of top-down, centralised organisations, be on the way out?  Satalia is just one example of a new type of decentralised, peer-to-peer company. It operates without managers and administrators at the inflection point between AI and organisational design.

Its COO (still a traditional job title) Avida Hancock explains: “We decided to architect what we think of as a new operating system for organisations, by using our technology to power the way we work, to enable people to collaborate in a completely decentralised, self-organised, purposeful way.

"We use data science and machine learning to extract insights from [our data] and to power a system that provides information to everyone in the organisation: what opportunities are available, what activities people are working on, who is connecting with who, how much projects are costing us, and our priorities within a collective strategy. This enables us to work in a completely non-hierarchical way.”

She adds: “A hierarchical construct is completely detrimental to an organisation... Yet we’re programmed from birth, in our education systems, to work in this way.”

Social platforms reinforce the point that status has a different meaning today: we update our ‘status’ daily, by telling others what we’re doing, rather than who we are in the social order. Is this the key lesson for innovating in 2018? Only you can decide.

About the author
Chris Middleton
See full profile

Chris Middleton is one of the UK’s leading business & IT journalists and magazine editors. He is founder of Strategist magazine, consulting editor and former editor of Computing. He is also the former editor of: Computer Business Review (CBR). He is the author of several books on the creative use of digital media, and has commissioned, edited, and/or contributed to at least 50 more. Unusually, Chris is one of the few private individuals in the UK to own a real humanoid robot, which he hires out to schools, colleges, and corporate clients. Robotics and AI are now core areas of Chris’ journalism.

Latest from Chris Middleton

To enable comments sign up for a Disqus account and enter your Disqus shortname in the Articulate node settings.

Recommended articles

Manifesto: The Innovation Economy

Smarter, safer drilling: innovation in the oil and gas industry

Humans and Technology: The Hybrid Generation

Black Mirror: Science fiction or a chilling reality?

Agtech: Where field and server farms meet

Latest articles

Why automation could lead to mass unemployment and how a Universal Basic Income might be the only solution

CERN’s asset management supercollider

5 Tips on How to Innovate

Process Model Innovation: A vital part of Modern Dual Innovation

Change activism for the really agile enterprise


Science and technology are the principal drivers of human progress. The creation of technology is hindered by many problems including cost, access to expertise, counter productive attitudes to risk, and lack of iterative multi-disciplinary collaboration. We believe that the failure of technology to properly empower organisations is due to a misunderstanding of the nature of the software creation process, and a mismatch between that process and the organisational structures that often surround it.