Four priorities for European business leaders navigating the AI era

Full text

Business in the age of artificial intelligence (AI) moves with dizzying speed. More powerful models launch regularly, bringing new opportunities and risks. Fresh use cases emerge daily, increasingly leaning on the orchestration power of agentic AI. Innovation boundaries recede as the cost of inference declines and robotics accelerates. It’s as if we’re permanently on fast forward.

I reflected on the speed of change with Samantha Garretto at Politico recently . A year on from our first meeting, the debate on AI in Europe has become more grounded, more urgent and more practical. There is now much more awareness of what AI can do, its limits and how to make it work.

But as AI develops at pace, the goalposts shift, meaning Europe’s leaders must accelerate in four areas: sovereignty, scale, security and sustainability.

Sovereignty: Go hybrid

Geopolitical tensions are disruptive — but can also be a catalyst for change. Conflicts in Ukraine and the Middle East, combined with flux in international norms, have sharpened calls for Europe to develop greater strategic autonomy — especially in energy, defense and technology.

On technology, Europe must continue to develop its AI ecosystem across the full stack, building on the strength of global champions such as ASML in hardware; SAP in software; and industrial AI leaders such as Siemens and Dassault. We have 19 EU AI factories already operating across the continent and plans to build up to five EU gigafactories.

Yet, while building a competitive ecosystem is critical, European organizations want options, not isolation. Our research suggests around a third of company workloads are sufficiently sensitive to require more control over data, models and infrastructure . The lower risk remainder need access to the scale and innovation muscle global players provide. That is why a hybrid approach is key and, we believe, should be reflected in the proposed EU Tech Sovereignty Package.

Scale: Support the long tail

Twelve months ago, I advised European business leaders to board the AI train — or risk being left behind. Encouragingly, we’ve seen a clear evolution from experimenting to scaling AI across the enterprise. Productivity retains primacy, accounting for ~50 percent of our work with clients. However, a quarter of engagements are now focused on using AI to boost topline growth, up from just 15 percent a year earlier. This mindset shift is underscored by our inaugural AI Progress Barometer , which tracks how quickly the world’s largest 3,000 organizations are building the capabilities required to scale AI.

We found that European companies are behind their North American counterparts but gaining ground, particularly in terms of upskilling and process reinvention.

I hope the gap narrows when the next edition of the barometer drops in November, but this will require two things. Firstly, broader adoption. The barometer reveals a large AI readiness gap between Europe’s larger and smaller companies, far larger than in North America. Our ‘long tail’ of smaller companies face specific capital, skills and tech barriers when adopting AI. Indeed we have just launched Accenture Edge to provide solutions that are faster to deploy, more repeatable and right-sized for the mid-market.

We found that European companies are behind their North American counterparts but gaining ground, particularly in terms of upskilling and process reinvention.

Secondly, we must address the tools versus talent investment imbalance. Too often capital is funneled into infrastructure while culture change and middle-management incentives are neglected. An idea I discussed with CEOs recently in London is to treat workforce training as an investment in productive capital — as we do with R&D and infrastructure — rather than an operational cost. That’s why, at Accenture, we invest ~$1 billion annually to train our people.

Security: Move from reactive to proactive

Recent debate around access to frontier models marks an inflection point in the development of AI. Boosting governance of new, more powerful models — such as those that could cause “systemic risk”, per the EU AI Act — is now firmly on the agenda across the globe.

However, there is only so much regulation can do in an interconnected world. AI is accelerating the scale, frequency and sophistication of breaches; 72 percent of organi z ations report a rise in cyber threats . The onus is on business leaders to invest more in security. Our recent research found that just 10 percent of companies globally have the strategy and capability needed to defend against modern, AI-driven threats.

Sustainability: Maximize tokens per kWh

Disruption in the Strait of Hormuz has sent shockwaves through energy markets. In Europe, natural gas prices are up almost 50 percent from the end of February , while storage sites are expected to refill to just 76 percent of capacity by October — their lowest level since 2011 heading into winter.

Meanwhile, increased AI use puts pressure on the energy grid; electricity consumption from European data cent er s is expected to increase by 49-187 percent to 2030 . We must therefore avoid the perverse incentives of ‘tokenmaxxing’ and, rather, take a more frugal approach that seeks to maximize the tokens per kWh. Using ‘right-sized’ AI models that might be smaller or industry-specific , employing smarter infrastructure and decarbonizing data centers can help.

Speed is the new currency

The pace of change is undoubtedly a challenge. From conversations with CEOs across different industries, I know many feel overwhelmed with the pace of change. But this is not business as usual anymore; this is the moment to accelerate business transformation.

The good news is that, with clean data and a well-defined use case, the time it takes to move from AI pilot to scaled solution is compressing. Across industries, our clients see the return on investment crystallizing: financial crime and ‘know your customer’ in banking; autonomous supply chains and demand forecasting in consumer goods; digital twins and predictive maintenance in automotive; and citizen services and contact centers in public services.

From conversations with CEOs across different industries, I know many feel overwhelmed with the pace of change. But this is not business as usual anymore; this is the moment to accelerate business transformation.

Indeed, such is the momentum that I am more confident about European competitiveness today than I was a year ago — even in the difficult macroeconomic context — because the conversation has changed. Twelve months ago, many organizations were still trying to understand the potential of AI. Today, the focus is on how to implement it in a way that both captures value and strengthens Europe’s position.

As AI continues to accelerate, Europe’s competitiveness hinges ever more on the speed of execution.