Artificial intelligence: How to build the business case

AI might be a hot topic but you'll still need to justify those projects.
Written by Mark Samuels, Contributor

VIDEO: Artificial intelligence's reach is soaring in 2017, but are businesses ready?

There's plenty of excitement around artificial intelligence -- analyst Gartner places it at the top of its top 10 strategic technology trends for 2017. The analyst says the technology has reached a tipping point and AI is beginning to extend its tentacles into every service, thing, or application, and that it will become the primary battleground for technology vendors looking to make money through 2020.

Interim CIO Christian McMahon, who is managing director at transformation specialist three25, acknowledges interest in AI has exploded recently, but he also voices a word of caution.

"All the major corporations, accelerators, and venture capitalists are desperate to find a foothold," he says. "However, I don't think the current AI market is at a stage where breakthrough technologies are about to be unveiled. Rather, it's a vibrant market which seems more conceptual than one of tangible substance."

It is a sentiment that chimes with Omid Shiraji, interim CIO at Camden Council. His organisation holds a huge amount of data and aims to use its knowledge to help people with complex needs. AI could provide a breakthrough in data insight, yet Shiraji says CIOs must focus on value creation.

"The business case for these projects is not easy -- you can take a step into the unknown," says Shiraji. "You sometimes have to rely on intuition rather than ROI to place your investments in these types of projects."

Gartner suggests executives who take a risk on AI projects will be rewarded and should consider experiments in one or two high-impact scenarios. So how will pioneering organisations build a business case for AI? Two IT leaders -- one each from the private and public sectors -- give us their take.

Sizing up the opportunity

Matt Peers, CIO of global law firm Linklaters, draws a parallel between the use of big data and the growing importance of AI. Peers says success in big data is all about being able to make the best use of the information you possess -- and Linklaters, a 175-year-old firm, is a business with more knowledge than most.

Peers says his organisation should be able to turn its history into a competitive advantage. Lawyers need knowledge about legal precedent, previous projects, and internal skills specialisms. He believes advances in AI will help his firm to create more sophisticated approaches to search.

"The key to success is getting the right information to people quickly," he says. "Some of the tools that are being developed for AI will help us search big data. Most of the technologies on the market today are good at clustering and reading contracts, and enabling you to search vast volumes of data for legal themes."

He expects the ability to digitize and search contracts for key legal themes to become commonplace very quickly. Linklaters has already created an AI working group to help analyze services in the marketplace and to work out how these technologies might impact the business.

"Firms in some key sectors are already making a move," says Peers. "We've spent a lot of time in the past 18 months sizing up the opportunities by talking to people, seeing demonstrations, and running proof of concept studies."

Peers recognises AI could also help change the way lawyers work, yet he also expects a cultural challenge. Senior partners trust their associates to spend hours considering the details of legal documents. Trusting computers to undertake the same task in seconds presents a different form of dependence.

"It's a big shift because the reputation of that lawyer and firm is on the line," he says. "The acceptance of AI in the business is going to involve an evolution. It's important to remember that there are many matters in the legal world where AI is not going to be useful for quite a long time. It's going to take a while for computers to provide trusted advice and opinion."

Using data to save lives

Toby Clarke, interim head of IT at Moorfields Eye Hospital NHS Foundation Trust, says AI will have a huge impact on the work of publicly-funded organisations. Moorfields has been working closely with DeepMind Research, a project that involves the Trust sharing a set of one million anonymised eye scans.

The project between Moorfields and DeepMind relies on historic scans, meaning that while the results of the research might be used to improve future care, they will not affect patients today. However, the hope is that discoveries through the initiative will lead to earlier detection and help reduce preventable eye disease.

"What they're doing with that information is truly amazing," says Clarke, referring to the DeepMind project. "It's cutting edge and will make a significant difference."

He says the key to long-term change through AI is being able to use information to inform patient care. And that use presents challenges, particularly in terms of data security and confidentiality. "The real value will come from using non-anonymised data," he says.

"If you have a large repository of information, and you can add big data from demographics, you can start to take make predictions about patient healthcare. You could potentially say when people should be coming in for tests in terms of early warnings."

The current project uses anonymous data. "It has to be that way," says Clarke. "In terms of healthcare, there will always be issues around how you commercialise data, and how you deliver value back to the host organisation and its patients."

Clarke, however, is keen to point out that similar projects could sponsor significant change. "It's difficult for humans to understand the impact of AI right now but the potential is huge," he says. "The technology self-learns and I find it exceptionally exciting. AI is different and new, and it's something everyone involved in IT should be investigating."

In contrast to reports that automation simply leads to job cuts, Clarke says AI -- particularly in the role of predictive medicine -- could lead to a whole new range of data science roles. "It's not about removing jobs but it is potentially about saving lives," he says.

Read more about business concerns for artificial intelligence:

Editorial standards