I was at the Thinknet6G Summit in October, which was an eye-opening opportunity to see what applications are under way for the next generation. I had the fortune to host a discussion and asked a panel of the assembled leaders what would change (other than maybe an icon on a phone) which would signal to them that 6G had come of age? What would it feel like to live in a 6G environment?
The answers were very smart, but very diverse… as might be expected, given the variety of objectives and outcomes talked about for the next generation. And that got me wondering – what are we doing if people neck-deep in 6G development can’t agree on what the outcome is? It reminded me of two people who have already signalled some alarms, not perhaps about the objectives the industry’s working towards but about the process.
Ken Figueredo of More With Mobile already touched on this change in some 2024 research. The telecoms industry has moved from one where, in 2G, operators had a very clear set of requirements they required vendors to meet in order to achieve a particular global objective; then the vendors worked to the requirements and competed on that. There’s a very clear command-and-control process which global standards were well suited to.
In 6G, though, governments set out an array of sustainability and societal targets for ubiquity, digital equity and so on; researchers and vendors talked about various applications and capabilities such as ISAC; and telcos were very clear that they wanted ideally some kind of incremental and software-based set of upgrades that didn’t cost much. The clear command-and-control process had changed.
Meanwhile, consumers were no longer the mass-market customer of a device and a data plan. Instead, enterprise services of all kinds would need to be unlocked. But where, with consumers, a ‘build it and they will come’ mass-market approach was effective, it’s become generally accepted that that won’t work with the varied needs of enterprises. So what does that mean for the process?
Figueredo suggested that the value chain should now be driven by end-user enterprise requirements, which the telecoms ecosystem should gear up to support. Instead, we have a situation where global scale still requires standardisation, but those standards are largely driven by a relatively small set of vendors. The telcos may adopt them if it makes financial or business sense to them, but in 5G so far only a minority of the standards have seen commercial application.
In other words, the process overall has broken down because the process tries to answer a different kind of problem to the one we face today.
Wicked
What kind of problem do we face? Credit to futurist Caroline Van Holden, who spelt it out at 6GSymposium in May 2021.
6G development is what’s known as a wicked problem. There are 10 traits which identify a wicked problem, and 6G development meets all of them (skip over this next bit if you’re willing to take my word for it).
There is No Definitive Formulation. The 6G vision encompasses contradictory and evolving objectives. For example:
- Economic competitiveness vs digital equity and inclusion
- Massive coverage vs massive capacity and lower latency
- Innovation velocity vs stability and interoperability
The “No Stopping” Rule. When is 6G “complete”? The answer is inherently unclear:
- Commercial deployment target is 2030, but full ecosystem development extends far beyond
- Requirements evolve as use cases emerge during development
- Debate still rages about whether 6G is ‘the last G’ or not
Solutions are Better/Worse, Not True/False. No objectively “correct” 6G architecture exists, or is likely to:
- Spectrum allocation choices range hugely
- Debates over centralised vs. distributed intelligence, and what any of that would look like
- Arguments over the role and relationship of cellular, Wi-Fi, Satellite etc; likely to look very different from country to country.
There is No Immediate Test. The consequences of 6G design decisions won’t be fully known until:
- Large-scale deployments occur in the 2030s
- Security vulnerabilities are discovered in production
- Energy consumption impacts become measurable
- Economic and social equity effects manifest
- Geopolitical implications unfold
Every Attempt Counts. 6G standards decisions have irreversible implications:
- Spectrum allocations by the ITU affect decades of infrastructure investment
- Architecture decisions in Release 21 lock in design patterns
- As the patent landscape solidifies it affects competitive dynamics
There’s No Enumerable Set of Solutions. A practically infinite set of technical possibilities exist across:
- Physical layer waveforms and modulation schemes
- Network slicing implementations
- AI/ML integration approaches
- Space-air-ground integration architectures
- Security and privacy mechanisms
- Implementations of functionality to do with energy-saving, ISAC or other services
The problem is Essentially Unique: 6G is unlike any previous generation. It is:
- The first to explicitly target sustainability as a core objective
- The first to integrate AI/ML as fundamental, not supplementary
- The first to plan satellite/terrestrial integration from the start
It’s a Symptom of Larger Problems. 6G development difficulties reflect deeper systemic challenges:
- Global and national digital divides and inequality
- The climate crisis and sustainability imperatives
- Geopolitical competition and technology sovereignty
- Cybersecurity and national security concerns
- An environment where vendors hoard most of the deep technical understanding
There are Multiple Valid Explanations. Different stakeholders frame the 6G “problem” differently:
- Operators need revenue growth and operational efficiency
- Vendors want market opportunities and strong competitive positioning
- Governments want national security, economic competitiveness, and digital sovereignty
There is no “Right to Be Wrong”. The stakes are extraordinarily high.
- $100B+ infrastructure investments globally
- Environmental implications (energy consumption, e-waste)
- Social implications for privacy, surveillance, and digital equity
- Geopolitical power dynamics and technological leadership
Anti-Wicked Behaviour
It is possible to respond well to wicked problems. The three basic principles for doing so are well understood and fall into the APT acronym.
Firstly, Adaptability is critical. Any good ‘solution’ is liable to create new questions or challenges, while a bad one will worsen the situation. As a result, iteration, feedback and learning are essential. Essentially it’s replacing command and control with directed trial and error. The faster you can iterate, and the smaller steps taken in each iteration, the better. It places less risk in any one iteration and lets you judge whether it’s closer to, or further from, being a desirable outcome.
Which leads on to the next principle: Participation. Who decides what a desirable outcome is? In a situation with many stakeholders, the more of them that are engaged throughout the process the better. That way they can feed in and advise on each iteration – “Yes, that’s more what we want” or otherwise. It means that development is directed in a way which creates demand and responds to the market. We don’t get situations where a technology or service comes to market and nobody’s interested.
However, the big caveat here is that it means building a shared understanding across different stakeholders and enabling meaningful participation from would-be end users. This is a very different proposition from what we’re used to, but unless the telecoms development process builds in meaningful participation then we are always going to be gambling with what goes out into the market – “build it and we hope somebody will come”.
The third principle is related to that: being Transdisciplinary. The aim is to integrate knowledge from a wide range of domains while developing; not just technical, but social, economic, environmental, and political domains. Co-creating solutions and knowledge offers a much better understanding of what the endpoint is, and what the ramifications of actions are.
The point is that technology does not stand alone in the real world. Its adoption is influenced by many other factors and, in turn, it affects these other domains for better or worse. By engaging with these other domains in tackling a wicked problem, better and more desirable outcomes can be found.
Is that to say it’s impossible to achieve a good outcome any other way? Not at all. But it does become much more a matter of luck than judgment.
The Horror
Any insiders from standards organisations are likely reading this and laughing hysterically at the idea of bringing in yet more viewpoints to the discussions they have. The 3GPP process of consensus development is already hugely difficult, and the people managing it are right to be proud of how they manage it. What kind of consensus are they liable to have if they bring in non-technical people as well, or people who have vastly different skill sets?
In many ways, what has been a strength – consensus – could easily turn into a weakness if unchecked. Inclusive processes can reach better, more legitimate decisions than expert-only processes, but we need to distinguish between giving input and making decisions. The broadest possible base should give input, while other roles would be in positions to take that input and translate it into actions, whether in terms of trials or expanding upon trials to create standards or de-facto workable approaches.
The problem is that a set-up like 3GPP works very well in directing technical capability towards achieving a particular technical goal. That’s all very well when the goal is clear and it is a question of technology.
I think it’s hard to argue that there is a specific technical problem which 6G aims to address, though. Instead there is a wide variety of societal, political and business goals which 3GPP is being tasked with finding a technical solution for. It’s like giving somebody a hammer and asking them to rewire a house with it; they can give it their best shot, but it’s just the wrong tool for the job. It’s going to leave a mess and ultimately the fault lies with the person asking, not the person with the hammer.
So what other models exist, and can we learn from them?
Anti-Horror
Firstly, an honourable mention to SNS-JU, who have an approach that pools pre-competitive research, explicitly aims to solve societal goals, brings in perspectives from other industries and runs large-scale pilots for iterative trials. It’s much closer in organisation and spirit to those ‘anti-wicked’ APT principles. However, while SNS-JU can feed into global standards, it is limited in how far it can influence the final direction.
However, to see an organisation that’s really operating against a wicked set of problems, we should turn our attention to the world of agriculture.
Climate change, population migration, changing demographics, wealth inequality, political unrest, epidemiology, changing water availability, ecosystem change – all of these create a shifting set of problems which are global in scale but different in impact from place to place. Finding ‘a solution’ to feeding people is like trying to nail water.
So what does the intergovernmental group CGIAR (Consortium of International Agricultural Research Centers) do which is so different?
Firstly, they work with a genuinely diverse set of stakeholders – from governments and NGOs, scientists and agricultural vendors through to the farmers and their supply chains). Their objectives and governance are based on outcomes and impact, which means that ultimately they care about things like crop yields, smallholder income, and which they measure transparently on a results dashboard. This moves discussions away from technical deliverables which are only the means to an end, even though the organisation is explicitly research-based.
Second, CGIAR works explicitly with a “Theory of Change” methodology. This means that they map how interventions create outcomes across different contexts, and then they revise their work and initiatives based on that evidence. What is more, while they conduct generic research at a global level – for example, on developments with genetics – they use regional bodies and partners to implement that in a specific context.
Finally, the structure of the organisation means that there are very many inputs, but people are playing to their strengths. It’s not one assembly where research scientists are debating with farmers and governments, but there are feedback mechanisms to ensure that science is directed in ways which can be applied effectively. Meanwhile, there is transparency in where funding comes from, where it goes to and who is accountable for what.
We could also learn lessons from the IPCC – the Intergovernmental Panel on Climate Change. Again, they are dealing with a wicked problem. In their case, their work runs with a slightly different transparency and engagement mechanism.
The IPCC reports and recommendations work in a very specific manner. Among the original drafting teams geographical and gender balance requirements need to be met, which forces the basic thinking away from addressing problems in ways which are relevant only to specific regions. Their work explicitly highlights and quantifies uncertainty. They use a language of likelihood to specifically acknowledge unknowns, which removes any implied suggestion of infallibility and encourages inputs which might otherwise not be forthcoming.
They will typically start reports and programmes among specialists, and their recommendations or reports then go for progressively wider review and redrafting; first to governments, then to the wider public. This enables any stakeholders to have a voice and to engage with the problems the IPCC addresses, but it also compels those drafting the reports to use language which is understandable to their readers.
We might argue that standards specifications are openly available for anyone to read, but how many people have the time and expertise to wade through 1000-page technical documents and understand them? It’s a rare skill and tends to be hoarded by companies involved in setting the standards. In mediaeval Europe reading Latin – and therefore the Bible – was a skill largely reserved for priests, who became as a result the intermediaries between the populace and their religion. We have a similar set of intermediaries today in telecoms.
The Missing Link
While the telecoms industry is one of the most organisation-rich, with partnerships and bodies for almost everything and every group, today we have no parallel for bodies like CGIAR. We might argue that some aspects of the GSMA and other fora work similarly to IPCC, but even then we’re probably stretching a point.
Instead, we have organisations such as NGMN, GSMA, ETNO, TMF and others who support (usually national-scale) telecoms providers; technical standards groups such as IEEE and 3GPP; groups supporting particular technologies or approaches such as WBA, O-RAN Alliance, AI-RAN Alliance, LoRaWan Alliance, the Bluetooth SIG and many others; and groups engaging with particular parts of the market like the CCA, GSOA, GSA and others.
What we don’t have is any group looking at how the industry evolves, or responsible for managing that process effectively. Without that, we hand over responsibility for good outcomes to the standards bodies because they are the nearest thing we have… or else, as is increasingly the case, other groups spin up to answer specific questions in parallel with, or in contradiction to, work going on in standards. None of which deals with the most wicked question of how the telecoms industry as a whole can address the challenges set before it.
Can we create one? Is it maybe a function for the ITU or the GSMA to take on? Perhaps we build a group from the ground up, or by expanding the remit of SNS-JU.
I don’t pretend to have the right answers here… but maybe this is a good (and not TOO wicked) question to start from in finding one.
