Parallel Births, Different Purposes

The Internet of Things and Artificial Intelligence share a curious relationship: both emerged from entirely different intellectual traditions, yet find themselves inseparable in the modern enterprise. To understand where we are today, we must trace where each began.

The term “Internet of Things” was coined by Kevin Ashton in 1999 while working at Procter & Gamble. Ashton used the phrase in a presentation connecting RFID technology in P&G’s supply chain with the internet. The idea was deceptively simple — give physical objects a digital voice. For business, this was revolutionary. Instead of humans manually logging inventory or reading meters, machines could communicate information themselves. [RFID Journal]

Artificial Intelligence has older, more philosophical roots. The 1956 Dartmouth Summer Research Project on Artificial Intelligence is widely regarded as a foundational event in the field. Organized by John McCarthy and other researchers, the meeting helped establish artificial intelligence as a distinct area of research. [Dartmouth]

From Curiosity to Core Infrastructure

IoT found early business traction because its value proposition was often tangible and immediate. When a manufacturer could see where a pallet was in real time, or when a utility could read meters remotely, the potential operational value was relatively easy to understand. Adoption was often driven by operations teams looking for efficiency and visibility.

AI’s enterprise journey was more complicated. Early business AI included rule-based expert systems, but widespread adoption accelerated as cloud computing, large datasets and increasingly powerful computing resources made machine learning more practical. Work on deep learning in the mid-2000s helped contribute to that resurgence. [Hinton Research]

IoT gave businesses a nervous system — the ability to feel and sense the world. AI gave them a brain — the ability to interpret and act on what they sensed.

The evolution of intelligent enterprise systems

The contrast in adoption patterns is instructive. IoT expanded horizontally across industries — logistics, agriculture, manufacturing, healthcare and retail — with individual use cases often focused on monitoring, connectivity and automation. AI increasingly expanded into the decision layer, using data to identify patterns, make predictions and automate increasingly sophisticated tasks.

A Tale of Two Timelines

Their evolutionary paths differ across several dimensions — from how they generate value to how organizations structure adoption.

Sources: IoT Analytics · Gartner · NTT / WSJ Intelligence · Dartmouth

Where Each Technology Changed the Game

IoT’s early impact was especially visible in manufacturing and logistics. Connected sensors can monitor equipment, inventory and operating conditions, while edge computing allows data to be processed closer to the devices generating it. These capabilities have helped make predictive maintenance, remote monitoring and real-time operational visibility practical at scale. [IBM]

AI’s impact has been especially visible in areas such as financial services, healthcare, retail and technology, where organizations have large datasets and complex decision-making processes. Machine learning and other AI techniques can identify patterns, make predictions and support decisions that would be difficult to perform manually at scale.

By the 2020s, the distinction between an “IoT company” and an “AI company” began to blur. Smart systems increasingly combine sensing, connectivity, machine learning and automation. John Deere’s See & Spray technology, for example, combines cameras, edge computing and machine learning to distinguish crops from weeds and make real-time decisions about where to spray. [John Deere]

Amazon provides another example of this convergence. Its fulfillment operations combine robotics, machine learning and computer vision to automate and optimize physical movement and order processing. [Amazon]

Neither Revolution Came Without Cost

IoT’s growth brought with it a significant security challenge. The proliferation of connected devices — including devices that may have weak default credentials or inadequate security controls — created new attack surfaces. The 2016 Mirai botnet demonstrated the risk dramatically, using vulnerable IoT devices to conduct a major distributed denial-of-service attack against DNS provider Dyn. [CISA]

AI faces a different set of challenges — many rooted in trust, governance and accountability. Bias can arise from datasets, algorithms, organizational processes and human decision-making. NIST identifies systemic, computational/statistical and human-cognitive forms of bias as risks that organizations should consider when developing and using AI systems. [NIST]

Both technologies also confront a common challenge: the skills gap. Organizations implementing connected infrastructure need people who understand networks, devices, data pipelines and security. Organizations implementing AI need expertise in areas ranging from data engineering and machine learning to governance, security and responsible deployment.

When the Two Technologies Became One

The most significant development of the last several years is not what either technology achieved independently, but what they can accomplish together. The term “AIoT” — Artificial Intelligence of Things — describes systems where AI capabilities are integrated into connected-device environments rather than existing only as a downstream consumer of IoT data.

Edge AI is one technical manifestation of this convergence: AI inference models can run directly on or near IoT devices, reducing latency and dependence on constant cloud connectivity. Edge computing allows computation to happen closer to the point where data is generated, which can improve speed and support applications where real-time decisions matter. [IBM]

For business strategy, this convergence represents a fundamental shift in competitive advantage. Companies that built robust IoT data infrastructure can potentially use those datasets to improve analytics and AI applications. Companies with strong AI capabilities can use connected devices as a source of real-time information about physical operations.

Generative AI adds another dimension to this story. Large language models can increasingly interface with enterprise systems, allowing people to interact with data and operational information through natural language. The result is a shrinking boundary between physical infrastructure and intelligent software.

The Business Imperative of the Next Decade

The question for business leaders is no longer whether to invest in IoT or AI — it is how to build the integrated infrastructure that makes both work in concert. The organizations that will be best positioned for the next decade are building what might be called the “intelligent enterprise”: an organization where data flows from sensors and systems, is processed and acted upon by AI, and feeds back into improving both physical and digital operations.

The divergence between these two revolutions is collapsing. Manufacturers that built connected factories can now layer AI-powered analytics and predictive systems onto that infrastructure. Agricultural companies are combining sensors, cameras, machine learning and automation. Retail and logistics organizations are combining connected inventory systems with AI-powered forecasting and robotics.

What began as two parallel revolutions — one rooted in physical connectivity, one in computation and machine intelligence — has converged into a single story about how businesses collect intelligence from the world and act on it faster, smarter and more continuously than ever before.

The future belongs not to IoT companies or AI companies, but to organizations that have made intelligence — physical and artificial — inseparable from how they operate.

The intelligent enterprise thesis