English translation of the Portuguese original. The audio is in Portuguese.
In the /root episode with Alexandre Uehara, published in 2023, I brought up the internet of things to start the conversation in a more didactic way. I asked whether IoT had reached the limit of its evolution or was only at the beginning.
Has IoT evolved over the last 10 years? Are we at the limit of IoT’s evolution, or only at the very beginning?
Alexandre’s answer — summary
Alexandre answered that every technology evolves, even when it seems not to. To illustrate, he recalled that in a short time the market went from a phase when everyone was talking about NFTs to another when everyone was talking about the metaverse, and that at that point the conversation had already moved to artificial intelligence.
His central point was the moment of each technology, when it truly meets a business model. On its own, he said, technology does nothing. The metaverse was his example: the concept already existed back in the days of Second Life, but the technologies it depends on were still missing. When the topic came back, it ran once again into the question of which business model would sustain it.
According to him, something similar happened with IoT. Sensors and connected objects had been in use for quite some time, but the surrounding technologies needed to improve these applications or turn them into a business were missing. By then, that had changed, and he cited 5G as an example of a technology that IoT works very well with. That is why he suggested looking at an ecosystem of technologies that, combined, meet the business model, instead of assessing each one in isolation.
A reflection for today
In 2023, Alexandre’s answer pointed to the ecosystem that was missing around sensors. Looking at it today, one piece that has gained ground is artificial intelligence running close to the machine, on the equipment itself or on an industrial computer next to it. Sensors no longer just collect data to send elsewhere; they feed a model that interprets what is happening right there.
Two Siemens announcements in 2026 help illustrate this shift. In March, the company introduced Drivetrain Analyzer Onsite, which analyzes vibration signals and other data from motors and drives with locally executed AI to identify deviations from normal behavior and early signs of wear, keeping the data within the customer’s own infrastructure. In April, it announced the general availability of the Industrial AI Suite, a platform for deploying and managing AI models on the shop floor in applications such as predictive maintenance and visual inspection.
These are product launches, and they do not say how widely they are already used or what results each operation will achieve. Still, they show a path I find promising for IoT: sensors, local processing and AI working together on a concrete problem, such as anticipating motor maintenance or removing a defective part from the line. For anyone evaluating a project, I would start with that problem and with the data that already exists, and only then choose the technology.
The full episode is available on YouTube.
Credits: /root — O Podstech da Celcoin. Hosts: Adriano Meirinho and Thiago Zaninotti. Guest: Alexandre Uehara. Episode 7: IoT: Humano x Máquina, published in 2023. Question at 1:43–1:52; summarized answer at 1:52–4:18.