What businesses need to fix now to avoid expensive 6G lock-ins | Computer Weekly

Why 6G is not just a mobile upgrade 

As a system-level shift, 6G is far more than simply an incremental mobile connectivity upgrade. Rather than treating AI-driven networks, edge computing, sensing and communications as add-on features, as with 5G, 6G is expected to fundamentally integrate them into core architecture. 

For enterprises, this could considerably change the role of connectivity, as data transmission signals will also be used for environment monitoring, motion detection and automation support, especially in smart traffic management, industrial robotics and emergency services.

6G will also introduce higher levels of network automation, with self-optimising networks handling real-time resource allocation, network configuration and inference management. Predictive maintenance is expected to help resolve traffic spikes and network failures before they happen as well, which will boost reliability.

As an “edge-native” architecture, 6G will have distributed intelligence and low latency, which can significantly advance remote surgery, augmented reality and on-device model training.

“The real value of 6G lies in device density, deterministic latency and integrated sensing, not headline gigabit rates. Despite early warnings about cyber security risks, many early discussions lack a strong emphasis on security by design,” says Elbehiery.

Basically, given the widespread impact of 6G on core infrastructure, treating it only as “faster 5G” could be a crucial mistake. 

Where enterprises are already getting 6G wrong 

By looking at 6G only as incremental connectivity, several enterprises are already making investment and infrastructure planning decisions based on near-term feature expectations that may not exist when practical deployments come into effect.

One of the most immediate mistakes is mis-timing commercialisation and investment strategies. By failing to devise early revenue generating and monetisation strategies and still relying on 5G’s “build and they will come” mindset, enterprises risk treating 6G as a frantic race. This can lead enterprises to over-investment in premature technology before standards are fully defined, rather than developing practical deployments for the 2030s.

Organisations could also make architectural and infrastructure mistakes. By relying on single vendors and closed architecture, enterprises will vastly increase the risk of expensive and complicated lock-ins down the line, especially since 6G standards are still in flux.

There are inflated expectations that 6G will fix 5G shortcomings such as dead zones as well. However, in reality, bridge technologies such as 5G-Advanced (3GPP Release 18+) will likely continue to be needed. Similarly, 6G is unlikely to eliminate the need for Wi-Fi through universal coverage, due to persistent practical constraints such as high frequencies struggling to penetrate walls. 

“Many organisations out there are deploying edge compute in a way that is optimal for current 5G use cases, without thinking about what those environments are going to need to do in terms of interoperability across several networks and locations in a 6G world,” says Tomas Novosad, consumer technology analyst and founder of Fibre in my Area. 

Organisations are also under-accounting for system complexity and governance gaps in 6G, especially when it comes to the massive integrations, high energy and hardware needs the technology requires. Additionally, AI-native narratives have been overhyped at times, particularly around early 6G applications. While standards are expected to heavily integrate AI for beam management and energy efficiency, it is still likely to remain optional for critical infrastructure in the future.

With the rise of AI-manipulated network management, data ethics, privacy and accountability questions will become more complex and need to be addressed as well. 

The real danger: lock-in risk 

Currently, the biggest risk of early 6G decisions is lock-in, which can come in many forms. The most common is supplier lock-in, when enterprises continue to rely on established infrastructure providers to reduce integration failures. However, this can often result in high-cost, multi-year maintenance contracts and dependence on single ecosystems, which can be slow and complicated to reverse once deployed.

Cloud infrastructure highlights a similar problem, where data, networking and orchestration layers become less portable over time, embedded in single environments. When this happens, migrating to another provider becomes a full architectural rebuild rather than a quick technical transition.

Despite increased policy support, enterprises are still hesitant to fully embrace Open RAN, mainly due to commercial pressures. This often means that supplier diversification remains largely theoretical at scale.

Another risk is committing too early to pre-standardised 6G technologies (pre-3GPP Release 21). This can lead to infrastructure being built on assumptions that later standards do not hold up, necessitating expensive replacements or redesign once standards are fully determined.

Early architecture and spectrum decisions may also cause long-term rigidity at the physical layer. 6G is likely to need much more mid-band spectrum than 5G. However, committing infrastructure to current assumptions, could cause incompatibility with potentially higher-frequency and denser deployments years later.

Given the accelerating shift towards a multi-network future, which will include 6G, Wi-Fi, satellite systems and private networks, overinvestment in any single network model could trap organisations into structures that no longer align with how connectivity is delivered. 

“The most expensive 6G mistake will not be buying the wrong radio. It will be building a network that cannot evolve to support innovation and the explosion in service demand without a procurement crisis,” warns Leid Zejnilovic, co-academic director of the digital data design institute at Nova SBE. 

What enterprises need to fix for 6G right now 

To minimise the chances of expensive 6G lock-ins and long-term rigidity, organisations need to take some concrete steps right now. “First, audit where current vendors control data gravity and operational workflows,” says Zejnilovic. “Second, insist on open interfaces for telemetry, policy and automation in new contracts. Third, put governance around AI-driven network changes in place now, before those tools become too embedded to challenge.”

A key step is building architectural resilience through network-agnostic, hybrid connectivity design, prioritising cost control over optimisation-driven investments. This requires modular, AI-aware architectures that separate connectivity from application logic and use portable interfaces to enable flexibility across both local and global environments.

Ideally, systems should be able to work across 5G, 6G, Wi-Fi and private networks, rather than focusing too narrowly on only one. This is because no single connectivity layer is expected to remain dominant or stable long enough to back long-term infrastructure decisions.

Private networks and edge deployments should be treated as useful targeted tools, not default architecture, as overuse can cause more fragmentation and operational complexity than they solve. This is particularly when they are deployed without clear workload-dependent criteria.

Organisations must prepare for automated and AI-driven networks as well. Observability, governance and accountability will all become much more critical as network operations become more autonomous. This is likely to shift risk from connectivity failure to hard-to-audit automation layers. 

“The question is no longer just whether the network can optimise itself,” Zejnilovic adds. “It is whether the enterprise can explain, constrain and reverse those optimisations when they affect performance, resilience or compliance. Auditability and human override will become core control points.”

Similarly, enterprises should avoid chasing incremental headline speed improvements and investments on the back of supplier-driven “6G-native” narratives, to minimise chances of future expensive rebuilds once standards are fully formed. 

From 5G to 6G: an uneven transition

Instead of a clean shift, the path from 5G to 6G will likely be messy, uneven and highly fragmented, impacted by geopolitical, commercial and technical limitations.

One of the main reasons for this is 5G’s ongoing underperformance in several areas and not being fully monetised, despite significant infrastructure investments. This has made funding for yet another massive infrastructure upgrade for 6G harder to obtain currently. As a result, rather than a reset, the technology is being seen more conservatively as an incremental extension.

Increasing focus on 5G-Advanced as a bridge to 6G has further complicated the transition. Instead of a clear shift, networks are expected to evolve in overlapping phases, which could extend a hybrid environment and slow large-scale adoption. 

Significant technical challenges – such as building much denser infrastructure and increasing energy capacity – remain. Both deployment complexity and costs are likely to be higher than previous generations.

Geopolitical fragmentation is shaping 6G development too. Digital sovereignty bids and competing standards across the US, Europe and China could hinder interoperability, creating a multi-speed global roll-out instead.

Most importantly, 6G still doesn’t have a clear commercial driver, unlike previous generations. There is no “killer app” or immediate need that supports widespread adoption, making demand uncertain, despite expectations of advanced sensing and holographic communication.

These factors highlight a transition which will be defined by coexistence, not replacement. As such, organisations need to be prepared to operate across overlapping connectivity generations, rather than anticipating a single, clear transition to 6G.

What 6G readiness looks like

6G isn’t about early adoption, but rather about avoiding potentially regrettable and expensive decisions today. Disciplined, architecture-first enterprises which focus on hybrid connectivity and network-agnostic systems will be most able to adapt as standards become defined.

However, organisations waiting for a clean reset and swayed by overhyped “6G-native” narratives risk losing more in premature investments that will be difficult to reverse down the line.

“6G is more than just another generation of wireless technology; it is a redefinition of how digital systems connect, compute and coordinate across the planet and beyond,” Elbehiery concludes. “Organisations that recognise this early will design for flexibility, openness and intelligence. Those that do not risk locking themselves into architectures that cannot evolve with the future.”

The real challenge is not preparing for 6G but learning to adapt to a messy, overlapping transition.