If you believe everything you see, you could easily believe we’re moving into a world where both the connectivity and the intelligence IoT relies upon are undifferentiated propositions. The most appropriate of these resources is selected automatically by an autonomous agent so there’s no need to form an opinion or preference about what your deployment uses. The outcome is that IoT becomes agnostic about connectivity and intelligence.
Recent innovations and introductions mean previously convoluted and complex processes for provisioning IoT connectivity and setting up intelligent resources at the edge or in the cloud have been resolved. The careful decisions of the previous decades, assessing connectivity types, providers and contract terms, are about to be replaced by an always-optimised automatic system.
That vision means your Cat-1 LTE is as good as your non-terrestrial network (NTN), your low earth orbit (LEO) satellite coverage or your low power wide area network (LPWAN) technology. It also assumes an uninterrupted migration from 4G to 5G to 6G and radically oversimplifies the sophistication of IoT connectivity and intelligence. You didn’t have to scratch the surface market spiel at the show to uncover barely-hidden realities in which multiple technologies will co-exist in an ecosystem composed of different providers who are yet to work out they will interact with each other in the automated world.
The clean vision of IoT resources provisioned according to need, through automated commercial agreements, supported by agentic AI, with flexibility assured isn’t really happening. No one really believes one human in a co-working space in Hyderabad or Helsinki is going to be finally responsible for your million-unit deployment with autonomous systems trusted to deliver optimised, secure connectivity whatever the situation.
What people at the show believe is that the new SGP.32 specification will make it easier for IoT devices to access connectivity profiles and that gathering more data into a single pane of glass (SPoG) management system will aid management precision and help handle massive scale.
There are differing views of what constitutes that SPoG and how comprehensive it should be. Is a SPoG that only covers IoT connections sufficient? Should a SPoG really provide granular information on connectivity, intelligent resources, the machine data and the context with adjacent devices? These are open questions which MWC26 visitors are engaging with. Visitors also believe that security can’t be left to chance, that connectivity is complicated and it should be treated with respect, and that agents simply aren’t ready yet.
That’s not to say IoT that is network and intelligence-agnostic won’t happen, just that it won’t happen soon or in the format that the over-excited suggest today. In the meantime, AI will be used to automate some processes and to rationalise data into SPoG displays, leading to faster, more accurate decision-making and more efficient IoT operations.
That, in itself, is a substantial innovative leap for IoT organisations to take and should be celebrated.
The author is George Malim, the managing editor of IoT Now.
Comment on this article via X: @IoTNow_ and visit our homepage IoT Now
