BUSINESS HQ MONTHLY

You don't need to be in Silicon Valley to succeed in AI

AI-powered smart grid system visualizing solar and wind energy management with glowing network icons, renewable power optimization and sustainable infrastructure concept. <i>(Image: Getty Images)</i>
AI-powered smart grid system visualizing solar and wind energy management with glowing network icons, renewable power optimization and sustainable infrastructure concept. (Image: Getty Images)
This article is brought to you by our exclusive subscriber partnership with our sister title USA Today, and has been written by our American colleagues. It does not necessarily reflect the view of The Herald.

There is a tendency when new technologies emerge to assume success belongs only to whoever builds the biggest version of it. Artificial intelligence is no different.

Much of the public conversation has focused on vast large language models, eye-watering valuations and a race between global technology giants. It can create the impression meaningful participation in AI is reserved for those with billion-pound budgets and hyperscale infrastructure.

I don’t believe that is true.

For businesses in Scotland and beyond the real opportunity is not to replicate Silicon Valley; it is to apply AI intelligently to the sectors, strengths and specialist expertise we already have.

That should come naturally to us. Scotland has never depended on being the biggest player in the room. Our history is richer than that. We have long punched above our weight through ideas, ingenuity and practical invention – from James Watt refining the steam engine, to John Logie Baird pioneering television, to the towering scientific mind of Mary Somerville, who advanced the celestial mechanics of Isaac Newton. Our advantage has often been intelligence applied with purpose.

That same mindset matters now.


Read more:


It starts by recognising an important distinction: bigger is not always better.

Large language models are remarkable general-purpose tools. They can summarise text, generate content and respond convincingly and conversationally across thousands of topics. However, many business problems are not general-purpose problems. They are specific, high-stakes, operational challenges that require precision, trust and context.

We should also be realistic about where this market is heading. We should not wait for large language models to eventually have all the answers, because they won’t. No single system will hold every nuance of every sector, regulation, workflow or commercial reality. The world is simply too complex, too fast-moving and too specialised.

That is why specialism is becoming the new moat.

The enduring value in AI may not sit with whoever has the biggest model but with those who combine trusted data, domain expertise and focused systems to solve real problems better than anyone else. Proprietary workflows. Sector knowledge. Tools tuned for one job and judged by outcomes, not spectacle.

For Scotland that is encouraging news. We are strong in industries built on expertise rather than hype: engineering know-how, financial judgment, energy systems knowledge, life sciences research, food and drink craft, advanced manufacturing and logistics. These are exactly the ingredients specialist AI needs.

In those situations smaller specialist models can often be the better answer. A simple analogy: you don’t go to the supermarket for dental surgery. You go to a dentist.

Likewise, businesses should not assume the largest publicly available AI model is automatically the best or most secure solution for every task. If you are trying to detect fraud patterns, assess eligibility criteria, route procurement opportunities, forecast maintenance needs or interpret regulatory documents, a focused model trained for that purpose may be faster, cheaper and more accurate.

This matters because Scotland’s economy is rich in exactly the kinds of sectors where specialist intelligence has value.

Financial services, engineering, renewable energy, life sciences, food and drink, advanced manufacturing, logistics and tourism all generate complex workflows, repeatable decisions and large volumes of information. These are fertile conditions for applied AI.

Scotland’s economic strengths were built on know-how, craft and problem-solving. AI works best in exactly those environments.

Take energy. Scotland has serious ambitions in the transition to net zero. AI can support grid forecasting, predictive maintenance, planning optimisation and supply-chain efficiency.

Take financial services. AI can help identify anomalies, streamline onboarding, support risk analysis and reduce manual administrative burden.

Take manufacturing. AI can improve scheduling, maintenance cycles, quality control and stock management.

Take food and drink. AI can strengthen demand forecasting, logistics and waste reduction across globally recognised brands.

None of those examples requires a peppy chatbot blowing smoke up the proverbial. They require smart systems solving difficult problems.

That, in my view, is where Scotland can excel.

There is another misconception worth challenging: that AI’s value comes primarily from replacing people. In most serious business settings, the opposite is true.

The strongest outcomes tend to come when AI amplifies human expertise rather than substitutes it. An experienced engineer supported by predictive tools becomes more productive. A finance professional using anomaly detection can focus on judgment rather than trawling spreadsheets. A skilled operator with better data can make faster decisions.

Technology creates leverage. People create accountability.

That balance matters because AI still carries real risks. One of the most under-discussed is the illusion of competence.

Modern AI systems can sound highly convincing while being partially wrong, outdated or entirely mistaken. Fluency should not be confused with understanding.

For business leaders that means caution is sensible – but paralysis is not.

The answer is not to ignore AI nor to deploy it blindly; it is to adopt it responsibly. Choose use cases where outcomes can be measured. Use systems grounded in trusted data. Keep humans in the loop for consequential decisions. Prioritise tools that are auditable and fit for purpose. In short: be commercial, not theatrical.

Scotland has a habit of underestimating itself in moments like this. However, our track record tells a different story. We have world-class universities, deep technical talent, entrepreneurial resilience and globally relevant industries. We have built breakthroughs before and we continue to build high-value businesses now.

More importantly, we understand how to solve practical problems with limited resources – a useful discipline in any technology cycle.

The next generation of AI winners may not be the loudest companies or the ones with the biggest models. They may be the firms quietly helping businesses save time, reduce waste, improve decisions and unlock growth.

That is a game Scotland can absolutely compete in. We do not need to build the biggest AI in the world. We need to build the smartest.

Rab Meliani is founder and chief executive of AI firm Ailsa

Get involved
with the news

Send your news & photos