Artificial Intelligence (AI) and the Internet of Things (IoT) are two of the most impactful and far-reaching technology developments of our time. Increasingly these two technologies are deployed together and the term ‘AIoT’ has come to the fore. In this article we discuss the market potential for AIoT devices and the AI Use Cases found onboard these devices.
1.1 Introducing AIoT
IoT devices can generate significant amounts of potentially valuable data and AI-analytics can be applied to this data in many locations, including ‘cloud’ data centres, various network edge locations, and on board the actual IoT devices themselves. All of these approaches have the potential to unlock significant value and new insights from IoT data.
When we refer to ‘AIoT’ we mean the deployment of AI Use Cases on board IoT devices.
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.
1.2 The market potential for AIoT devices
The potential for AIoT is significant. As illustrated below, Transforma Insights forecasts that total AIoT connections will grow from 1.4 billion at the end of 2023 to 9.1 billion at the end of 2033. This is a more than 6-fold growth in 10 years, resulting in a CAGR of over 20%. Overall AIoT represents a significant market with net additions growing from less than half a billion in 2023 to just over 900 million in 2033.
For context, we forecast a total of 39 billion IoT devices at the end of 2033, up from 16 billion at the end of 2023. Accordingly, as illustrated below, we forecast that 9% of IoT devices have on board AI in 2023, rising to 23% in 2033. The rate of growth of AIoT penetration of IoT devices slows towards the end of the forecast period, primarily due to the AIoT penetration of key consumer IoT applications reaching saturation.
Clearly, AIoT is not equally well-suited to different IoT applications and rates of adoption of AIoT will vary widely across different IoT Application Groups.
In the case of AV Equipment, for example, we expect that by 2033, approaching 4 billion of a total of 5.6 billion IoT devices will have AI capabilities onboard and so be AIoT. Similarly, a very significant share of Vehicle Head Unit and IT Infrastructure IoT devices will be AIoT. Conversely, a negligible (or zero) share of Inventory Management & Monitoring, Building Lighting, and Track & Trace IoT devices will have AI capabilities onboard.
As is the case with IoT, consumer devices dominate AIoT although the share does decline, falling from 85% in 2023 to 81% in 2033. This is mostly a result of a simple underlying dynamic, which is that the addressable markets for devices like smart watches, smart TVs and smart speakers (i.e. individuals and households) are far larger than the addressable market for, say, autonomous agricultural vehicles.
The adoption of AIoT in cross-vertical IoT devices (supporting applications that are similarly applied across all industry verticals, such as, for example office HVAC systems) rises from 7% to 10%. Meanwhile, the share of AIoT in enterprise markets is 8% in 2023, growing to 10% in 2033 as AIoT becomes deployed in support of ever more enterprise contexts, albeit with an adoption lag compared to simpler and higher-volume consumer markets. Within enterprise contexts, the biggest enterprise sectors are Government, Health, and Transport, and Finance, primarily driven by the adoption of CCTV, Healthcare Monitoring, Road Fleet Management, and Usage Based Insurance respectively.
1.3 The AI use cases found onboard AIoT devices
The concept of AIoT does not exist in a vacuum, and AIoT capabilities are only ever deployed onboard an IoT device in support of an AI Use Case that the device in question is required to support. Again, the influence of consumer markets is clear with the most frequently featuring Use Cases being Natural Language Processing, Chatbots & Digital Assistance (both often associated with Smart Speakers) and Image Processing & Analysis (often associated with Smart TVs).
Many AI Use Cases do, however, find traction beyond the core IoT applications that the casual observer might naturally associate with the AI Use Case in question. For instance, whilst Natural Language Processing might most obviously be associated with (A)IoT applications like Smart Speakers and Headphones, the same Use Case will also find traction in environments such as Vehicle Head Units and Televisions (in both cases to support voice control, for example).
Accordingly, we might from the perspective of a vendor of AIoT Natural Language Processing capabilities (either hardware or software, or associated functions), identify different kinds of opportunity in AIoT contexts. From this perspective, ‘Primary Opportunities’ and ‘Extended Opportunities’ could be defined as follows:
- Primary Opportunities are AIoT devices that have specific AIoT Use Cases as part of their primary function. For example, AIoT enabled Smart Speakers will almost always have AI Natural Language Processing capabilities.
- Extended Opportunities are AIoT devices that may host specific AIoT Use Cases to support enhanced functionality. For example, AIoT Televisions will quite often have AI Natural Language Processing capabilities.
From this perspective, and based on Transforma Insights’ AIoT forecasts, the total AIoT market opportunity for Natural Language Processing capabilities in 2030 is made up of 1.8 billion devices that are Primary Opportunities and a further 1.7 billion devices that are Extended Opportunities. Similarly, the total AIoT addressable market for the Image Processing & Analysis Use Case is forecast to be 1.4 billion Primary Opportunities and a further 2.5 billion Extended Opportunities.
1.4 Summarising the AIoT opportunity
AIoT merges the concepts of AI and IoT, enabling AI use cases directly on IoT devices to unlock benefits including better performance, enhanced compliance and reduced costs. The market is set to grow from 1.4 billion AIoT connections in 2023 to 9.1 billion by 2033. Key use cases include Natural Language Processing and Image Processing & Analysis, especially in consumer applications although adoption in enterprise markets is accelerating.