Governance in the age of AI

ARTIFICIAL intelligence (AI) is advancing at breakneck speed. New models, applications and business use cases are emerging almost weekly. With billions being poured into AI development globally, the technology is no longer a niche or novelty – it is fast becoming central to how businesses operate and compete.

Governments are also rapidly embracing AI. Singapore’s National AI Strategy 2.0 is a case in point. It aims to leverage AI to transform the economy, boost productivity across various sectors and create new areas of growth. This includes strengthening Singapore’s AI ecosystem and developing a skilled AI workforce.

For board directors, AI is no longer just an IT issue. It has become a governance priority, across both across performance and conformance aspects. From ethics to operational risks, boards now have a key role in overseeing how AI is adopted and managed across their organisations.

How should boards be thinking about incorporating AI for the boardroom?

“Most valuable player” in the boardroom

AI’s superpower lies in its ability to leverage vast amounts of data – market trends, financials, customer behaviour, competitor intelligence – and serve up sharp and actionable insights. It allows boards to simulate scenarios, weigh strategic options and identify risks or opportunities more comprehensively than before.

This is why AI is becoming the new MVP (most valuable player) in the boardroom. It brings speed, depth and breadth to decision-making in ways few technologies have before. This would strongly complement the traditional management papers and presentations.

Routine tasks, too, are being transformed. Meeting transcripts, report summaries and schedules can now be handled by AI, saving time and improving efficiency.

AI can also flag compliance anomalies and irregular transactions, helping boards spot trouble early. Predictive tools are beginning to shape decisions on resource allocation, growth strategies and risk management.

Implementation challenges

The promise of AI comes with caveats. The current state of AI technologies may still falter when it comes to nuance, industry context or domain-specific language.

Sometimes, AI “hallucinates”, confidently giving answers that sound plausible but are completely inaccurate. With better training and improved technology, this will likely change. For now, however, boards must treat AI-generated insights as helpful – not gospel.

Another challenge lies in keeping teams AI-ready. More data does not automatically lead to better decisions if those interpreting the information are not equipped to understand it.

Boards must ensure their organisations invest not just in the technology, but in the people and processes required to use it effectively. They must guide the AI transformation with a clear technology strategy and the right long-term partners.

There is also the risk of outsourcing too much thinking to machines. Mental arithmetic has faded with calculators, and GPS has dulled our sense of direction. Similarly, AI could erode critical thinking if relied on too heavily. Complex, ethical decisions still require human judgement, empathy and context – qualities that AI does not yet possess.

Operational risks must also not be ignored. Data breaches, unreliable vendors and system failures are serious concerns. These developments raise difficult questions about oversight, safeguards and control.

Regulators are also racing to keep pace. Singapore’s Model AI Governance Framework and AI Verify toolkit represent important efforts to promote responsible use.

Other jurisdictions are developing their own responses – the European Union’s AI Act is one example. However, a clear and unified set of global regulations, laws and ethical guardrails remains some distance away, and would almost certainly continue to lag behind technology advances.

The governance gap

AI’s rapid ascent has exposed significant gaps in governance. Many boards are still playing catch-up, often without the frameworks or expertise needed to manage this evolving landscape. The result is real exposure to legal, ethical and reputational risks.

Take confidentiality. Using AI tools during board discussions about strategy or legal matters could inadvertently expose sensitive information. Worse, some AI-generated records may be discoverable in legal proceedings – even if they contain errors. In sensitive contexts, this could result in the loss of legal privilege or a breach of regulatory compliance.

Bias is another longstanding issue. AI systems can inherit and amplify assumptions embedded in their training data. The lack of transparency in AI decision-making – the so-called “black box” problem – makes it difficult to identify and rectify these biases or errors.

Perhaps the most complex question is accountability. If a board relies on flawed AI analysis and makes a costly decision, who is responsible – the board, management or the technology provider?

The chain of accountability utimately rest with the board as the apex leaders of the organisation. This is precisely why boards must remain closely engaged, ask the right questions and ensure that strong guardrails are in place.

No turning back

The era of AI is no longer approaching – it has already arrived. Boards must learn to navigate this new environment with both openness and caution. We have evolved rapidly from the computerisation era to the digitalisation era and now the “intelligentisation” era.

Embracing innovation is essential, but it must be accompanied by sound governance, ethical foresight and human judgement.

AI has the potential to transform decision-making, enhance operational effectiveness and unlock strategic value. However, without thoughtful oversight, it can just as easily create new vulnerabilities. In an era where systems are becoming increasingly intelligent, the role of human wisdom is not diminished – it is more essential than ever.

The writer is a member of the Governing Council of the Singapore Institute of Directors.