For most of the assembled experts, the fact that IoT has seen a simmering of excitement surrounding it can only be a positive.

“The fact that the hype is over is a good thing in general, mainly because it hypes prices, it hypes what you can do but also it hypes people's ability to achieve it, which means that not only are the values of the companies going up, but the estimate time to market are decreasing,” explains Aurore Belfrage, head of early stage at EQT Ventures.

“Both these two components make it harder to adapt what you're building to reality, and as an entrepreneur myself, it’s really hard to build a business, and it's really hard to build a business within this IoT with a hardware component space as well. I'm loving the fact that the hype is over because that lets us get onto business.”

However, for some the hype hasn’t completely gone, but has instead transitioned and evolved.

“I think the hype is still on. The hype is just moving a little bit,” counters Patrice Slupowski, VP of digital innovation at Orange.

“I think the core of activity has physically moved from a rule of general 'everyone's doing it' to the ones that are actually willing to take the risk and prepare to understand what the market looks like,” adds Belfrage.

For those entering the market, however, there are challenges, particularly as the technologies surrounding IoT remain in flux.

“In B2B, the decision cycles are very long, and when people have to decide about deploying a lot of material, choosing a network technology, for instance, or even when they are to save some money or try to bring new services, they are reluctant to make a bad choice,” says Slupowski.

“There are a few topics like security, interoperability and even some topics on network choice, technology choices etc that are reasons for not going at that time. This means that they want to do some pilots and then they want to do a larger pilot and when they're really sure maybe they can quote something.”

Slupowski recognises this in his own industry, which despite being one of the fields most naturally positioned to adopt IoT technologies, remains relatively slow to act.

“From the telco standpoint, we're on the road of deploying at least connectivity and then platforms for customers and we know that those things, they are taking cycles of five years in order to study, standardise, start to plan deployment and then deploy the technologies,” he explains. “IoT is going to be enormous and very interesting for all of us.”

Even when companies do decide to take the plunge and get involved in IoT, they face significant issues that have not yet been fully resolved.

“We've heard that in some of the projects that people are undertaking, 50% of the effort goes into harmonising the data from different centres and putting it all together,” says Garner. “The data area is a mess as far as we can see, and really difficult to get going in a good way. “

This is complicated by the fact that organisations – and cities looking to implement IoT solutions in particular – often perceive their problem as unique to them, and so seek custom solutions rather than off-the-shelf alternatives.

“Most of the cities I know all around the world still think they have one problem and they're the only one to have it,” says Slupowski. “Every city thinks that they're the only one to have traffic problems, access to public facilities and they're absolutely the only one, so they need their platform with their name on it.”

When everyone is generating different platforms to resolve their issues, this produces a fragmented scenario where there can be multiple incompatible solutions to what are essentially the same problems. And as a result, the IoT space is increasingly seeing the proliferation of a host of competing platforms that prevent datasets being easily cross-referenced or combined to generate insights, working against the overall goal of IoT.

This will, in some form or another, need to be resolved, either through some form of intervention or natural market forces.

“Some people with IoT draw parallels with the roll-out of the electricity grid – the electricity of things as it would have been in the mid-1920s – and there were I think two or three major contenders with how to do national electricity, and it was only when the government stepped in and said 'we're doing that one', that's when it was rolled out in a big way,” says Garner.

“Is it the government's role to do that in IoT? Because it's not that we have a lack of standards, it's almost that we have too many, don't we?”

This would ensure that standards were more rapidly agreed upon, but it comes with its own issues.

“Governments shouldn't have to mandate an IoT specification, especially in the UK, because what happens if every other country has got something else and we're left behind?” asks Jonny Voon, innovation lead at Innovate UK.

“For me it's more about software-defined autonomy. For example, Facebook shut down two of its AI bots because they developed their own language between them because it was more efficient, so why couldn't we spin that around, use intelligent machine learning or software-defined autonomy to determine what is the best way to communicate with other things in this vicinity?”

“You're back in the same problem. Which machines? Is it my machines that get to determine, or is it his machines that get to determine?” counters Belfrage.

“I don't know the answer to the question, but one of the things that we debated a lot is the timing of these things. Maybe there should be a time of chaos where there is a struggle, and the benefits of that, and when would the timing be right for a broader standard?”

In some industries, such as telecoms, standards are conventionally agreed across the industry before seeing a uniform rollout, something that is currently being planned for 5G mobile networks. However, Slupowski does not believe this is yet possible in IoT where the technology is still finding its feet.

“On the IoT we cannot say 'stop it, stop your IoT project, we need to standardise that',” he says. “The situation now is that you have a few guys trying to develop standards, and every time they meet together they're representing two standards, we have another one come in because of their meeting.”

Instead, he argues, we need to ride out the “chaos time” that IoT is currently experiencing, while encouraging solutions best suited to the clients at hand, be they cities or companies. And in time, the technology will mature enough for standardisation to emerge.

The IoT has a long road ahead if it is to realise its potential, but while many of the technological and structural issues will be resolved by time and market maturity, there is an issue of morality that still poses significant unresolved questions.

“I'm a firm believer that we will solve the hurdles, but what are the implications of that and where will that happen?” asks Belfrage. “I personally don't believe that the smart city, the AI development, will be equally distributed across everyone on the planet. We will see the powers of centrifugality and network effect that will collate power and data and capital, and capital to be taxed, in hubs.”

Not only does IoT have the potential to exacerbate rural and urban inequality, but it also presents key moral questions over the handling of data.

“We're throwing technology, we're throwing acronyms at it, we're throwing hype words and buzzwords and money and everything else at these problems, when in reality it comes down to how we as people begin to deal with the influx of data,” says Voon. “A machine isn't going to be able to tell if this is ethically a good way to use the data or not. That's ultimately us.”

A key part of solving this issue, argues Slupowski, is maturity.

“We all agree on the long-term vision of the two parallel worlds, one being digital, the other one being real and tangible, but we're not agreeing on the roads or the time it will take to come to that vision,” he says. “And it's true almost everywhere, whether we're speaking about autonomous cars or have the ability to use all the possible bits of data that grow out generating things through the IoT.

“I think that we need a little bit of maturity and when it comes to data, all of the things which are done around data, we need maturity not just around computing but all the planning and use.”

Ultimately, however it is important to recognise that while IoT has potential now, there is far more to come, and the field will not reach maturity for quite some time.

“The vision is still there and is still just as shiny and glossy as it always has been, but there's much more realism and recognition about the journey we need to go on,” says Garner. “It's not going to just happen next week or next month or next year. It's a long way away.”