Do boards understand the potentially game-changing nature of AI or are they at risk of misunderstanding or misjudging the implications? We caught up with Ryan McCarthy, Board Leadership Centre Lead, to discuss what boards need to know and do when it comes to AI.

There can be no doubt that the AI era has arrived. Barely a year since ChatGPT landed with a bang, investment and innovation has poured into the sector.

Google has launched its powerful Gemini system and Elon Musk’s X has even introduced Grok, an AI modeled on the Hitchhiker’s Guide to the Galaxy. Spurred by the proliferation of AI tools, the EU has just agreed the world’s first laws to regulate AI.

Having come from virtually nowhere, AI must surely be on the agenda at every board meeting? Not quite, says Ryan McCarthy, senior audit partner with KPMG Ireland and head of its Board Leadership Centre, which works with directors to explore critical boardroom challenges.

I would say we’re not seeing AI discussed around board tables in a meaningful way yet,” says McCarthy, who has extensive experience with PLC and private company boards in an audit and advisory capacity. “When it does get discussed, it’s with the level of depth of discussing any new potential game changer – at the start you’re only scratching the surface.”

That needs to change says McCarthy, who believes every board member needs to “take it upon themselves to understand the issues and implications – don’t leave it to someone else.”

"Understand the issues and implications – don’t leave it to someone else."

Threat and opportunity

On the occasions AI does come up, the discussion invariably turns to the emerging threat or risk it poses, according to McCarthy. “What doesn’t get discussed yet is the genuine business opportunity. Threat and opportunity are not discussed in equal measure.”

AI also tends to be mentioned as an external factor that could affect the organisation rather than being grasped as an operational item to be examined. “Thematically, we are seeing more of the external, broad conceptual threat rather than ‘actually, here’s a real thing’,” says McCarthy.

“Based on the hype around AI, you would imagine there would be moments where someone has come into the boardroom and said ‘we have a new emerging threat and it’s this specific thing’. I haven’t really seen it in those terms. The discussions so far seem to have been at surface level.” 

Risk: rules vs principles

To explain that, McCarthy takes the debate back a step, to a broader point about boards and their approach to risk. He challenges whether boards have become focused on risk primarily from a corporate governance point of view, as opposed to a practical point of view.

The risk section in a typical annual report is getting “thicker and thicker” – not without reason, McCarthy notes - but it can contain a lot of ‘cookie-cutter’ risk. He ticks them off on his fingers: cyber-attacks, supply chain, climate change and more. Now AI is getting added. 

“There has been a steady drift over time towards rules rather than principles,” he says. “So people ask whether it is all written down and documented, as opposed to asking what’s really going to sink the ship. You rarely find a really good example where someone has identified a specific risk to the company and actually set out an explanation of it and how they mitigated it.”

Advising the experts

So, if the AI discussion is still skin-deep at boardroom level, what will move it onto the agenda? McCarthy points to the modern governance structure, which involves companies drawing on specialists in areas including audit, risk, nomination/remuneration matters and, more recently, sustainability.

“There is a challenge: who advises boards?” he asks. “Do they get the right external advice? Do they ask the right questions? In that regard, are boards now looking for the right advice and insights in terms of technical experts on AI, bringing in real case studies from other companies and industries? I’m not sure the answer is ‘yes’; in fact, I’m pretty sure the answer is ‘no’.”

Some companies, particularly in the US, have created the dedicated role of Chief AI Officer. For his part, McCarthy suggests there may be a gap for a technology or ‘emerging tech’ committee at board level. “There are requirements around having the right number of financial experts on a board; should every board now have a tech expert on it?”

"Do boards have people from different backgrounds in terms of skills, particularly technology? I would probably say that many don’t."

Diversity behind the boardroom door

That leads to a broader point: given the close correlation between youth and emerging technology, does the typical top-level boardroom have the right demographic to deal with AI? After all, it is still the exception to see young people on big company, PLC or semi-state boards.

“We have come a long way in terms of boardroom diversity but there is another layer to diversity that is getting exposed here: do we have young people? Do boards have people from different educational backgrounds, different backgrounds in terms of skills, particularly around technology and innovation? I would probably say that many don’t."

Outside the boardroom, a company’s executive - including the HR or Chief People Officer function - should also be getting to grips with AI. “If something like the AI opportunity is not coming up through the organisation to board level, then you’ve probably got to ask whether you have got an executive who are tuned-in,” says McCarthy. “In the same way that you need day-to-day skills to fully embrace the requirements of ESG, do you have the right skills in terms of AI?”

The workforce question

A lot of the narrative around AI involves technology ‘taking’ people’s jobs, particularly where they involve repetitive processes or services. Despite dramatic headlines, McCarthy says that conversation hasn’t really hit boardrooms yet – and it doesn’t need to be a negative one.

“What I haven’t discussed with any client is the AI opportunity in terms of workforce. Part of the reason is that people haven’t fully figured out use cases properly,” he says.

It looks like those use cases will become more apparent in 2024 and beyond. One Dublin hospital has begun using AI to assess radiology scans, while the national weather service has an academic collaboration to explore using AI and data science in weather and climate services. 

“Diagnostics is one of the clear cases for AI,” says McCarthy. "Before the pandemic the WHO was forecasting a shortfall of 18 million healthcare professionals by 2030 - and radiologists appear to be in really short supply."

“The medical profession is producing more and more diagnostic information yet there is a worldwide shortage of people who can review it. There is a whole load of sectors like that, where there is a shortage of skills or a skills gap. Is there a solution in the middle, with AI?”

McCarthy gives the example of companies with large customer service operations. Many of them have been through the cycle of using onshore customer service teams, to moving some aspects offshore, and then introducing bots; or some combination of all three.

“Now the question is being asked: is there another way? And I think we are only at the cusp of that other wave.” 

Easing the boardroom burden

One cohort McCarthy does see AI helping: over-burdened board members, who are lumbered with “enormous pre-reads” just days before a board meeting. The content and tone of a meeting can be dictated by the hundreds of pages of highly-detailed board papers, shutting down time for strategy discussion and looking forward.

“Are there opportunities in AI being used to help boards perform their function better, through more dynamic information provision?” asks McCarthy. “AI-supported dashboards could let board members see what’s going on in the business, rather than relying on the set-piece meeting.”

For example, a company’s internal audit department could use AI tools to monitor key financial information and make it available in real-time to the audit committee. In KPMG we have seen AI tools that are already available that will review a sustainability report and highlight inconsistencies or disclosure gaps, and potentially compare a company’s data to a set peer group.

“These tools can create room for the board to do what they should be doing - which is a really impactful discussion on any issues and remedies,” according to McCarthy.

"The best board members have an innate curiosity."

Curiosity is key

In any case, he certainly expects AI to feature more prominently in those boardroom discussions in future. “The best board members – and by extension, the best boards – have an innate curiosity,” he says. “There are focused, strong people on those boards who are just constantly curious.”

“There are two things in the world we should be curious about now: one is geopolitics and the second one is tech, but more specifically AI. If you’re a board and you’re not curious about those two things – and their impact on your business – there is something wrong.”

Get in touch

We understand the pressure boards and directors are under to get it right and to understand and to act upon the myriad of their obligations.

If you’d like to find out more about the support offered by the Board Leadership Centre we’d be delighted to hear from you. 

Read more from the Board Leadership Centre