AI is making headlines daily, with billions committed to infrastructure, research and massive AI models. But none of this is designed with manufacturers in mind.
Instead of getting lost in the noise, focus on developing a well-defined AI strategy built around specific challenges and operational goals. It won’t make the front page, but it will move the needle where it matters most: your factory floor. XpertRule’s Akeel Attar and Iain Crosley explain more.
The AI momentum that dominated 2024 has only ramped up in 2025. Major announcements from the UK, US, EU and China have tech giants and start-ups racing to dominate the field.
UK: The government unveiled its AI Opportunities Action Plan, aiming to position the UK as an “AI maker” rather than an “AI taker.” Backed by £14bn from several tech firms, the plan includes investments in data centers, AI research and pilot projects, ‘AI Growth Zones’ (to be announced in the summer), talent development and expanding the AI Safety Institute (AISI).
US: Shortly after taking office, US President Donald Trump announced the formation of Stargate, a joint venture between OpenAI (the developer of ChatGPT), Japan’s Softbank, Oracle and UAE investment firm MGX. The venture intends to invest $500bn in AI infrastructure over the next four years, including nearly a dozen data centers.
China: A week later, DeepSeek, a Chinese AI start-up, grabbed global attention with an open source, cost-effective large language model that challenges the assumption that American companies will dominate the AI market.
Europe: Most recently, Ursula von der Leyen, President of the European Commission, pledged €50bn, alongside €150bn from the private sector, to “supercharge” European AI innovation. Plans include upgrading EU supercomputers and building four AI gigafactories that will specialise in training very large AI models.
While these initiatives all promise economic growth through job creation, new innovations and further investments, their focus is heavily weighted toward infrastructure and foundational research, especially around large language models and generative AI.
These technologies are built for processing vast amounts of unstructured, human-generated data, making them a poor fit for the operational realities of industrial environments.
The Story Beneath the Headlines
It’s easy to get swept up by these sweeping announcements. But the devil is in the detail. And in most cases, details are conspicuously absent. Political and industry leaders talk in vague claims and abstract promises, a trend we called out in our previous article, Show Me, Don’t Tell Me.
What’s clear is that while $500bn investments sound impressive, they won’t lead to a surge in industrial adoption. AI built for consumer and enterprise applications is fundamentally different from AI capable of delivering value on the factory floor.
As Carl Ennis, CEO of Siemens UK and Ireland, recently noted, AI in manufacturing is “built on data generated by machines rather than humans.” This distinction is key. Large Language models (LLMs) process unstructured natural text; industrial AI models work with structured, real-time machine data captured directly from sensors, control systems and machine logs.
Moreover, manufacturers are increasingly cautious about where their data is processed. Instead of relying on cloud-based platforms, they are turning to edge analytics – keeping AI-powered decision-making closer to the production line for faster, more reliable outcomes.
Finally and more critically manufacturers want Responsible AI that is consistent, accurate, safe, auditable, and understandable.
Decision Intelligence = AI That Works For Manufacturers
For manufacturers, AI’s value doesn’t come from chatbots or Large Language Models. It comes from AI that delivers actionable predictive & decision models that address specific operational performance challenges. This is where Decision Intelligence comes into play.
Decision Intelligence (DI), use composite AI technologies that work with structured machine data captured directly from sensors, control systems and machine logs. These AI technologies include predictive machine learning (non LLM) for predicting machine performance and failures, Symbolic AI for capturing the expertise of plant engineers and operators, and Optimization algorithms to generate optimal machine control settings.
DI can address operational challenges – from identifying defects at speed, analysing machine performance to anticipate failures before they happen, to dynamically controlling machines to optimize performance. This level of responsiveness reduces waste, improves consistency and enhances agility – factors which directly impact your bottom line.
DI can bridge the gap between OT on the shopfloor (data collection, monitoring and control), and enterprise IT (Supply chain, procurement, sales). It equips teams with sharper insights to improve decision-making at every level. By integrating data across machines, systems and business functions, DI provides a single source of truth for operational intelligence. DI helps manufacturers make better, faster decisions in a world of constant variability – supply chain shifts, fluctuating demand and changing regulations.
When manufacturers deploy Decision Intelligence, they unlock powerful capabilities, including:
Set Yourself Up for Success
Adopting AI in manufacturing isn’t about chasing trends. It’s about applying Decision Intelligence in ways that deliver measurable results. To do that, focus on:
The Bottom Line
Despite the barrage of buzzwords, hype and flashy announcements, manufacturers must stay focused on what truly drives value on the factory floor. Headline-grabbing AI initiatives may capture public attention, but they rarely carry substance. And when they do, the AI promoted is often ill-suited to manufacturing.
While billions are being invested in the next chatbot, the real opportunity lies in AI tailored to industry’s unique needs, delivering practical, decision-enhancing outcomes. The winners in manufacturing will be those who apply AI with precision, using Intelligent Automation and Decision Intelligence to make better decisions, reduce waste and drive real business impact.
This article is part of our ongoing series, Reality Check: What AI Really Means for Manufacturing, designed to inform, inspire and help you implement AI effectively in your manufacturing operations.
Stay tuned for our next installment, where we break down the difference between strategy and strategic execution – a must-read for manufacturers ready to take decisive action.