State of IoT 2024: 18.8 Billion Connected Devices, Edge AI, and the Infrastructure Reality Behind the Growth Numbers

The global IoT device count hit 18.8 billion by the end of 2024 — a 13 percent increase over 2023 and a number that, on its face, suggests an industry barrelling forward with unstoppable momentum. But the headline figure conceals a more complex reality. Growth decelerated from 15 percent the previous year. Forecasts have been revised downward. And the three forces constraining deployment — cautious enterprise spending, persistent chipset supply bottlenecks, and a sluggish Chinese industrial recovery — show no signs of resolving quickly.

For enterprise technology leaders managing IoT deployments — or evaluating whether to expand them — the market data matters because it shapes everything from component availability and pricing to vendor roadmaps and integration platform investment. This analysis unpacks the latest market intelligence on connected device growth, connectivity technology shifts, the AI-IoT convergence, security imperatives, and what the 40-billion-device forecast for 2030 actually means for organisations making infrastructure decisions today.

18.8 Billion Devices — and Why the Growth Rate Matters More Than the Total

The trajectory from 16.6 billion connected IoT devices at year-end 2023 to 18.8 billion at year-end 2024 represents solid growth, but the deceleration from 15 percent to 13 percent year-over-year tells the more important story. IoT Analytics has revised its forward projections downward compared to its 2023 forecast, now projecting 40 billion connected devices by 2030 — still an enormous number, but reached on a shallower growth curve than previously anticipated.

The deceleration is not a signal that IoT adoption is stalling — 51 percent of enterprise IoT adopters still plan to increase their IoT budgets in 2024, with 22 percent expecting increases of 10 percent or more. Rather, it reflects the reality that macroeconomic headwinds, supply chain constraints, and geopolitical disruptions create friction that slows the conversion of planned deployments into installed devices. For infrastructure planners, the implication is clear: capacity planning should be built around the moderated growth trajectory rather than the more optimistic projections that circulated before 2023’s economic turbulence became apparent.

Table 1

Global Connected IoT Devices — Growth Trajectory (Billions)

Year Connected IoT Devices (Bn) YoY Growth Status
2022 14.4 Actual
2023 16.6 +15% Actual
2024 18.8 +13% Estimated
2026f ~24 ~13% CAGR Forecast
2030f 40 ~13% CAGR Forecast

Source: IoT Analytics, State of IoT Summer 2024. Devices counted as active aggregation nodes/gateways. Excludes computers, phones, tablets, and one-directional technologies (RFID/NFC).

Three Structural Headwinds Moderating IoT Growth

Enterprise spending caution persists. Economic uncertainty remained the dominant topic in CEO earnings calls throughout 2023 and into 2024. Even as inflation moderates and interest rates begin to stabilise, the psychological impact on enterprise investment decisions creates a lag effect — business confidence recovers more slowly than economic indicators, leading to a wait-and-see posture that delays IoT deployment timelines. The 51 percent of enterprises planning budget increases is encouraging, but it means 49 percent are holding flat or cutting — and even the planned increases are coming off a period of constrained spending.

Chipset supply constraints are structural, not cyclical. Semiconductor supply has improved from the acute shortages of 2021–2022, but lead times remain elevated compared to pre-pandemic levels. Government initiatives to expand domestic chip manufacturing — the U.S. CHIPS and Science Act, the EU Chips Act — represent massive investments, but the lag between policy announcement and production output is measured in years, not quarters. TSMC’s planned fab in Dresden, Germany, will not begin production until 2027 at the earliest. Until new fabrication capacity comes online at scale, chipset availability will continue to constrain the pace at which IoT devices can be manufactured and deployed — particularly for the specialised, low-power chips used in industrial and LPWAN applications.

China’s industrial recovery is slow and uneven. China’s economic challenges are having a disproportionate impact on industrial IoT. By mid-2024, approximately 30 percent of Chinese industrial firms were operating at a loss — a level not seen since the 1998 Asian financial crisis. Companies losing money do not invest in new IoT infrastructure. Meanwhile, U.S. sanctions on Chinese-made chip exports and global inventory de-stocking strategies have forced nearly 11,000 Chinese chip and module-related companies to cease operations in 2023 alone. China remains the world’s largest IoT market by installed base, but its capacity to drive incremental growth is significantly constrained by these structural pressures.

Nearly 11,000 Chinese chip and module-related companies ceased operations in 2023 — a consequence of global inventory de-stocking, U.S. export sanctions, and a domestic industrial downturn that has left 30 percent of Chinese industrial firms operating at a loss.

— Trusted IoT analysis of IoT Analytics data

Three Technologies Dominate Nearly 80% of All IoT Connections

The IoT connectivity landscape is dominated by three technologies that together account for approximately 77 percent of all global IoT connections: Wi-Fi (31 percent), Bluetooth (25 percent), and cellular IoT (21 percent). This concentration matters for enterprise IoT architects because it defines the infrastructure investments, protocol expertise, and vendor ecosystems that will deliver the highest return on deployment effort.

Wi-Fi leads at 31 percent of connections, with three-quarters of Wi-Fi-enabled IoT devices shipped in 2023 based on Wi-Fi 6 or Wi-Fi 6E standards — delivering the faster throughput, lower latency, and improved device density handling that IoT deployments in smart buildings, healthcare facilities, and industrial environments require. Wi-Fi 7, which began shipping in 2024, is expected to account for 7 percent of IoT-based Wi-Fi shipments and will further improve performance in high-density environments. Bluetooth holds 25 percent, driven primarily by Bluetooth Low Energy (BLE), which has become the default connectivity option for battery-powered IoT devices — smart home sensors, asset trackers, wearables, and increasingly industrial sensors through IO-Link Wireless technology. Cellular IoT — spanning 2G, 3G, 4G, 5G, LTE-M, and NB-IoT — accounts for 21 percent and grew 24 percent year-over-year in 2023, significantly outpacing overall IoT growth. The introduction of 5G RedCap in 2024, which prioritises affordability and reduced complexity over ultra-low latency, is expected to drive adoption in video surveillance, asset monitoring, and enterprise IoT applications that need more bandwidth than LPWAN but less than full 5G.

Table 2

IoT Connectivity Technology Market Share (2023)

Technology Market Share Key Use Cases 2024 Development
Wi-Fi (6/6E/7) 31% Smart buildings, healthcare, industrial Wi-Fi 7 shipping; 7% of IoT Wi-Fi shipments
Bluetooth / BLE 25% Asset tracking, wearables, smart home IO-Link Wireless for industrial adoption
Cellular IoT 21% Fleet mgmt, video surveillance, utilities 5G RedCap launch; 24% YoY growth
LPWAN ~10% Smart metering, agriculture, logistics LoRaWAN and NB-IoT expansion
Wired (Ethernet/Fieldbus) 4% Industrial automation, PLCs, gateways 33 end nodes per aggregation node avg.
Other (satellite, proprietary) ~9% Remote/maritime, proprietary industrial Satellite IoT gaining traction

Source: IoT Analytics, State of IoT Summer 2024. Market share based on aggregation node connections. LPWAN and Other are Trusted IoT estimates based on reported data.

The AI-IoT Convergence: Edge Intelligence Becomes the Defining Trend

If there is a single theme that defines the IoT market in 2024, it is the integration of artificial intelligence — and specifically edge AI — into IoT device architectures and data processing pipelines. AI was discussed in 34 percent of all corporate CEO earnings calls by Q2 2024, making it the most prominent technology topic in boardroom conversations. The generative AI market went from negligible to white-hot within a single year, and its implications for IoT are becoming concrete: Siemens launched its Industrial Copilot for manufacturing environments, and NVIDIA and AMD are both investing heavily in edge AI silicon designed to bring inference capabilities directly to IoT devices.

Edge AI fundamentally changes the IoT value proposition. Instead of IoT devices passively collecting data and streaming it to the cloud for analysis — consuming bandwidth, introducing latency, and requiring persistent connectivity — edge AI enables devices to process data locally, make real-time decisions, and transmit only the insights or anomalies rather than the raw data. For industrial IoT applications, this means predictive maintenance models running on the factory floor, quality inspection systems making millisecond pass-fail decisions at line speed, and autonomous systems that operate reliably even when cloud connectivity is interrupted. For enterprise IT architects, the shift to edge AI creates new infrastructure requirements: more compute at the edge, more sophisticated device management, and new model lifecycle management workflows that ensure AI models running on distributed devices are kept current, monitored for drift, and validated against production performance.

IoT Security: 400% Attack Growth and the Regulatory Response

The security picture is sobering. IoT-targeted malware attacks grew 400 percent year-over-year in 2023, with manufacturing accounting for 54.4 percent of all reported IoT attacks — making it the most targeted sector by a wide margin. The attack surface is expanding with every new connected device, and the asymmetry between attacker sophistication and IoT device security maturity remains a fundamental vulnerability across the market.

Governments are responding with regulation. The UK became the first country to mandate IoT cybersecurity standards, requiring that consumer IoT products meet minimum security requirements before they can be sold. The EU’s Cyber Resilience Act imposes similar requirements across the European market. The United States has established a voluntary labelling programme for wireless consumer IoT products — less stringent than the European approach but a signal that regulatory expectations are tightening. Two emerging security architectures are gaining traction as responses to the escalating threat landscape: post-quantum cryptography (PQC), which addresses the risk that advancing AI capabilities could crack current encryption algorithms, and zero trust security, which eliminates the assumption that devices inside a network perimeter can be trusted and instead verifies every access request individually. For enterprise IoT deployments, the combination of regulatory compliance requirements and the accelerating threat environment means that security architecture is no longer an add-on — it is a deployment prerequisite that affects device selection, network design, data handling, and ongoing operations.

Sustainability as an IoT Accelerant: From ESG Compliance to Automated Reporting

Sustainability and ESG compliance have become the fastest-accelerating driver of IoT adoption. Boardroom discussions about sustainability increased from 9.4 percent of earnings calls in Q2 2019 to 23.6 percent in Q2 2024 — a trajectory driven by proliferating climate and sustainability legislation across jurisdictions: the UK’s Sustainability Disclosure Standards, India’s Green Credit Rules, the EU’s Corporate Sustainability Reporting Directive, and others.

The connection to IoT is direct and increasingly mandatory. Corporate sustainability reporting requires data — energy consumption, water usage, emissions, waste volumes — that must be collected continuously, accurately, and at a granularity that manual measurement cannot achieve. IoT sensors provide the measurement infrastructure: energy meters tracking real-time electricity consumption, water quality sensors monitoring pH, temperature, and turbidity, air quality monitors measuring particulate matter and chemical concentrations. The regulatory pressure to report this data is converting sustainability from a voluntary corporate initiative into a compliance obligation — and compliance obligations create predictable, non-discretionary demand for the IoT infrastructure that automates data collection and reporting. For enterprise IoT vendors, sustainability-driven deployments represent one of the most resilient demand segments in the current market, because they are driven by regulatory timelines rather than discretionary investment decisions.

18.8B

Connected IoT Devices (2024)

400%

IoT Malware Attack Growth YoY

40B

Projected IoT Devices by 2030

34%

CEO Calls Discussing AI (Q2 2024)

Frequently Asked Questions

IoT Market, Connectivity, and AI — Common Questions

How are IoT devices counted — and why does the methodology matter?

IoT Analytics counts active aggregation nodes and gateways — not every individual sensor or actuator. A single industrial PLC controlling 100 field instruments counts as one IoT device. A gateway aggregating data from 50 temperature sensors counts as one device. This methodology produces the 18.8 billion figure. However, the total number of wired IoT end nodes alone reached 23.4 billion in 2023, with an average ratio of 33 end nodes per aggregation node. In factory settings, this ratio can exceed 1:100. The distinction matters because infrastructure planning needs to account for total data-generating endpoints, not just gateway-level connections, and the 40-billion-device forecast refers to aggregation nodes — the actual number of connected endpoints generating and transmitting data will be many times larger.

What is 5G RedCap and why is it important for enterprise IoT?

5G RedCap (Reduced Capability) is a 3GPP Release 17 specification that creates a new category of 5G device optimised for IoT use cases that need more bandwidth than LPWAN technologies (NB-IoT, LTE-M) but do not require the ultra-low latency and extreme throughput of full 5G NR. RedCap devices offer download speeds up to 150 Mbps, upload speeds of 50 Mbps, and latency under 100 milliseconds — at significantly lower cost, power consumption, and device complexity than full 5G modules. The target applications include video surveillance (where bandwidth needs exceed LPWAN capabilities), industrial sensor networks requiring moderate data throughput, wearable devices, and smart grid infrastructure. For enterprise IoT architects, RedCap fills a gap in the connectivity portfolio between low-cost LPWAN (sufficient for simple telemetry) and full 5G (overkill for most IoT use cases and too expensive for high-volume deployments).

What is edge AI and how does it change IoT architecture?

Edge AI refers to running artificial intelligence inference models directly on IoT devices or on edge gateways close to where data is generated, rather than sending all data to the cloud for processing. This architectural shift has three practical benefits: it reduces latency (enabling real-time decisions without round-trips to cloud infrastructure), reduces bandwidth consumption (transmitting insights rather than raw sensor data), and improves reliability (devices can continue operating intelligently during cloud connectivity interruptions). For IoT architects, edge AI introduces new infrastructure requirements — more powerful edge compute hardware (driving demand for AI-capable chipsets from NVIDIA, AMD, and Qualcomm), edge-native model management tools (for deploying, monitoring, and updating AI models across distributed device fleets), and new security considerations (protecting model integrity and preventing adversarial manipulation at the edge).

Why is manufacturing the most targeted sector for IoT cyberattacks?

Manufacturing accounts for 54.4 percent of all reported IoT cyberattacks for three interconnected reasons. First, the attack surface is enormous — modern factories deploy hundreds or thousands of connected devices (PLCs, SCADA systems, industrial sensors, robotic controllers) that were often designed for operational reliability rather than cybersecurity. Second, the consequences of disruption are acute — production stoppages cost millions per day, creating strong incentive to pay ransoms quickly. Third, the convergence of IT and OT (operational technology) networks has created new attack vectors — devices that were previously air-gapped from the internet are now connected to corporate networks and cloud platforms for analytics, remote monitoring, and predictive maintenance. The combination of high-value targets, weak historical security posture, and expanding network connectivity makes manufacturing the ideal target for attackers seeking maximum impact with minimum resistance.

How does IoT support corporate sustainability reporting?

Corporate sustainability regulations — the EU’s CSRD, the UK’s Sustainability Disclosure Standards, India’s Green Credit Rules, and others — require companies to report detailed environmental data: energy consumption, greenhouse gas emissions, water usage, waste volumes, and supply chain environmental impact. Collecting this data manually is impractical at the granularity and frequency that regulations demand. IoT sensors provide the automated measurement infrastructure: energy meters tracking real-time electricity consumption per facility, per floor, per production line; water quality sensors monitoring discharge compliance in real time; air quality monitors measuring emissions at source; and logistics sensors tracking fuel consumption and route efficiency across supply chains. The regulatory timeline is creating non-discretionary demand for IoT deployments — companies must collect this data regardless of their investment appetite, which makes sustainability-driven IoT one of the most recession-resistant segments of the market.

What is zero trust security and why does it matter for IoT?

Zero trust is a security architecture that eliminates the traditional assumption that devices inside a network perimeter can be trusted. Instead, every access request — from every device, every user, every application — is verified individually, regardless of whether it originates from inside or outside the network. For IoT, zero trust is particularly relevant because IoT devices are often deployed in physically accessible locations, may have limited built-in security capabilities, and increasingly bridge IT and OT networks that were historically separated. Zero trust implementation for IoT includes strong device authentication (verifying that each device is what it claims to be), micro-segmentation (isolating IoT devices into their own network segments to limit lateral movement if a device is compromised), continuous monitoring (detecting anomalous device behaviour that might indicate compromise), and least-privilege access (giving each device only the minimum network permissions needed for its function). The practical challenge is that many legacy IoT devices were not designed to support zero trust protocols, creating a transition period where older devices need compensating controls while the fleet is modernised.

Trusted IoT is an independent publication covering trends and developments in industrial technology and software. This analysis is based on publicly available data from IoT Analytics’ State of IoT Summer 2024 report and related market intelligence. Trusted IoT is not affiliated with IoT Analytics, any device manufacturer, or any company mentioned in this article. This content represents our editorial analysis and should not be construed as investment or procurement advice.


© 2026 Trusted IoT. All rights reserved.