AI: urgent transformation or considered choice?

Terry Storrar at Leaseweb UK asks how quickly businesses should be going ‘all in’ with artificial intelligence

In less than two years, AI has gone from a relatively niche technology topic to one of the world’s biggest talking points and the destination for huge levels of investment. Analysis by Goldman Sachs, for example, equates the growth of generative AI with “innovations in electricity and personal computers” with the potential to have a bigger impact on GDP than either of those historic tech trends.

Predictions about the impact of AI across just about every industry sector have become so prevalent that business leaders are under considerable pressure to identify how AI can deliver transformational benefits.

According to a study by EY, technology has become the number one cause of business disruption in 2024, up from sixth place 12 months ago and now ahead of talent, economic, geopolitical, climate and consumer/social factors. This unprecedented rate of business change has been “catapulted by advances in generative AI,” according to its authors.

Clearly, the potential of AI is enormous, but there is also a growing sense that organisations should focus on balancing the benefits of AI with financial efficiency before they go ‘all in’. For instance, the extent to which the average company can translate AI hype into deliverables remains open to question.

Real-world obligations

In this context, business leaders should carefully consider a number of important issues, such as whether investment in AI can deliver the levels of data security, flexibility and reliability they are going to need. In what is fast becoming a race to transform, every organisation should evaluate what groundwork is required before AI can be meaningfully deployed.

Take the issues associated with data readiness, for example. This is the process of preparing organisational data for use by AI models and is reliant on the proper storage, processing and management of data required for its efficient, accurate and secure use in AI/ML applications.

In particular, AI models are dynamic systems that will generally need to ingest and adapt to new data – many on a constant basis. As fresh data is added, the underlying hosting infrastructure must adapt to ensure adequate storage and network capacity is available at all times.

There are also sustainability issues to consider. Power-hungry AI applications can have a considerable impact on environmental performance, with the significant energy consumption required for training large AI models leading to higher carbon emissions. According to Goldman Sachs, AI is poised to drive a 160% increase in data centre power demand, with the average ChatGPT query requiring nearly ten times as much electricity as a Google search.

Clearly, every responsible organisation will need to factor this into their decision-making processes.

Investment in people

Building AI-centric organisations isn’t just about technology investment and implementation. As processes are redesigned or enhanced, there is likely to be a commensurate requirement for workforce upskilling and the development of tailored talent acquisition and retention strategies.

Business leaders need to ensure they have the skills within their teams to identify and act on the opportunity for innovation that AI offers.

While larger companies will have the option to accelerate their process with more rapid investment, smaller organisations must ensure they balance the acquisition of AI technologies with their ability to fully harness the benefits. It’s very unwise, for example, for a business to make a significant financial commitment to AI only to find they don’t have the ability to properly implement, support and build on what the technology has to offer.

Sitting behind all of these issues is the increasingly complex regulatory environment, with governments racing to catch up with the pace of change. Among the most important examples is the recently adopted EU Artificial Intelligence Act, a framework designed to ensure that AI systems are developed and used in a safe, transparent and trustworthy manner – the goal being to prevent harm and ensure fundamental rights are protected.

Compliance breaches could be met with large fines of up to €35 million or 7% of global annual turnover for banned AI applications and/or €15 million or 3% of turnover for violations of obligations under the AI Act. It’s important to note that businesses in the UK that develop or deploy an AI system used in the EU must comply with the Act.

Bringing all this together can help bring some pragmatism into the decisions organisations make about investing in AI. In many cases, the argument about whether to invest or not has already been won, but the point is, how far and at what speed should business leaders be going?

Those who put the correct groundwork in place, who build the infrastructure and prepare their teams for an AI-augmented future, will be much stronger placed to succeed in the long run. Without this approach, however, adopting AI could be a bumpy ride. 

Terry Storrar is Managing Director of Leaseweb UK

Main image courtesy of iStockPhoto.com