Transforma Insights, the world’s leading IoT analyst firm has today published a report looking at the key implications of the convergence of Artificial Intelligence and the Internet of Things, discussion of which has recently exploded under the banner of ‘Physical AI’ or ‘AIoT’. The report examines opportunities, impact on supporting infrastructure, potential speedbumps, and more.
Transforma Insights’ new report ‘AI and IoT: what are the implications of the convergence of the technology domains?’ examines the growing interest in the intersection of Artificial Intelligence and the Internet of Things, which has emerged recently under the concepts of ‘Physical AI’ and ‘AIoT’. The report focuses on 5 areas: AI as a driver for IoT, AI on board IoT devices, supporting infrastructure and cloud-to-edge orchestration, AIoT hardware, and the use of AI to support IoT operations.
One of the main reasons why AI is seen as being so interesting for IoT is that it will act as a driver for IoT adoption. Clearly, given AI’s ability to crunch large volumes of data, there are significant implications for removing some of the barriers to deploying IoT. Furthermore, Transforma Insights is of the view that there is greater value derived in AI at the point where physical and digital world intersect (i.e. where IoT operates), for instance in autonomous driving, defect detection, health & safety. AI-enabled IoT has the potential to create more tangible value than many purely digital AI applications, as well as reducing the cost of deployment. However, there are challenges. IoT is already a multi-disciplinary exercise involving hardware, software and connectivity. Adding AI into the mix isn’t going to make it any easier to roll out.
One of the key elements of the convergence of AI with IoT is the increasing deployment of AI onto the edge device itself, what we term ‘AIoT’. Applying AI to IoT data on board the source IoT devices can bring significant benefits, including improved performance, enhanced compliance, privacy and security and potentially reduced operational costs. The report examines the motivations for deploying AI onto IoT devices, examples of propositions, the complexities of managing it in the field and the need to balance edge and cloud processing.
In quantitative terms, based on our AIoT forecasts, we predict that total AIoT connections will grow from 1.8 billion at the end of 2024 to 11.3 billion at the end of 2035.
The advent of more AI on IoT devices has a number of implications for supporting infrastructure. It reduces raw data streaming to the cloud. It necessitates frequent updating and lifecycle management, often across heterogeneous fleets. It changes the nature of data storage architectures. It increases the importance of security and governance issues. It necessitates a change in approaches to traffic management. And much more. As a result, and as laid out in the report, there is an increasing requirement for new supporting infrastructure, most prominently what we term AIoT platforms, featuring a range of new functionality, as illustrated below.
The report also examines the implications for hardware with the increasing prevalence of AI-optimised hardware, the use of AI internally by IoT suppliers to support their operations, and considers some of the key challenges (‘speed bumps’) that will hinder deployment, including model management, security, resource constraints in IoT devices, technical complexity, and the challenge of integrating such deployments into business processes.
The report can be found here: 'AI and IoT: what are the implications of the convergence of the technology domains?' (subscription required).
The report is available as part of the Transforma Insights Advisory Service. Please get in touch with us if you would like to learn more about the service.
The topics covered in the report were also the focus of a recent Transforma Insights webinar, ‘AI and IoT: a marriage made in heaven?’. The replay of the webinar is available here.
