Often in this world, it takes time for revolutionary technologies to live up to their full potential.
Most of us know that gunpowder was first invented in Tang-Dynasty-era China, where it was put to use in fireworks. Fewer may know that the steam engine was first invested in ancient Greece, where it was used to power curiosities for the elite.
The “killer app” for both of these technologies came centuries much later.
Although the timescales involved are exponentially shorter, the same will be true of AI.
From office productivity tools to industrial revolution
So far, we’ve seen a large number of office-based productivity tools come to market. In the UK, the most popular workplace AI tools are Copilot (58%) and ChatGPT (48%), and the tasks most commonly addressed through AI summarising information (60%), research (58%), and editing and checking text (56%).
These are useful tools that will deliver important productivity lifts for businesses. But they’re not true “killer apps.” For that we need to look at bespoke, industrial use cases.
New data from IFS shows that the organisations currently pulling ahead are those that apply AI to the unique, specialised demands of their individual industry, rather than deploying generic AI solutions across the enterprise.
AI is entering a new phase, where industry-specific AI solutions will become the engine of growth. This is where the true potential of the technology will be realised.
Tailoring AI to industry needs
Every industry has its own needs. In the energy and utilities sector, for example, 48% of firms say that predictive maintenance is their top AI use case. Meanwhile, 41% of manufacturers say it’s supply-chain optimisation that delivers the most value. Twenty-eight percent of construction organisations focus on applying AI to financial planning and forecasting for complex build projects.
With so many differing requirements, it’s a no-brainer that a general AI application will be of less use than one that’s custom built for a specific industry and its needs. We are therefore entering a new phase of AI, where adoption is driven by fundamental industrial priorities.
Our research shows that currently 60% of businesses in industries such as manufacturing, energy and utilities, transportation, aerospace and defense, construction, and telecoms are in the process of deploying industry-specific AI.
These are AI solutions that combine contextual intelligence, governed digital workers, and advanced AI-decisioning to help industrial businesses run, optimise, and transform their operations. In these cases, AI does much more than simply generate text or summarise information. It acts within critical processes to reduce unplanned downtime, prevent waste, improve operational efficiency, and much else besides.
These deployments are already showing early signs of promise. Despite not yet being fully deployed, nearly 60% of industrial leaders say that AI has saved their teams between one and seven hours per week. Eleven percent report that over seven hours have been saved.
The race to scale
The next challenge is to expand these initial experiments and operationalise AI broadly across the industrial enterprise.
Here, many organisations are coming unstuck. According to our research, factors including data quality and accessibility (24%), a lack of clearly defined industry use cases (17%), and skills gaps (16%), security concerns (16%) and systems integration (16%) are obstacles on the journey to scale.
The industrial success stories of the future will be the businesses that understand how to overcome these challenges to connect data, define the right use cases, and embed industrial AI into everyday workflows.
In many cases, this will require a shift to a unified, industry-specific data model that natively connects ERP, asset management, and field operations. Meanwhile, overcoming skills shortages requires moving past generic AI toward embedded, domain-specific intelligence that connects with workers directly within their daily workflows.
Competitive advantage comes down to speed of execution
Approximately half of the industrial leaders we spoke to reported that they’re confident in their ability to execute on AI fast enough to remain competitive – leaving half who are not.
That’s concerning given that competitive advantage is linked to speed of execution. Businesses that can scale AI ahead of competitors will have a significant advantage in the years ahead. The rest may struggle to keep up.
It’s encouraging to see businesses getting to grips with the industrial applications of the technology. We’ve seen what AI can do to office productivity tools, and we’re now beginning to see the impact of the technology in areas including supply chain and asset performance.
However, it will only be when AI is scaled up and deployed more broadly and consistently that we will see the technology live up to its true revolutionary potential. Businesses that make this leap the soonest by unifying their data, mapping powerful use cases, and embedding specialised AI into industrial workflows will be the sector leaders of tomorrow.
