AI GUIDE FOR BOARDS IN SINGAPORE
Section I: The Board Mandate
• Data leakage and IP poisoning risks: As enterprises race to build proprietary agentic workows using corporate data, the risk of data ex ltration, model poisoning, or unintended intellectual property leakage via public or semi- private cloud infrastructure escalates sharply. • Systemic concentration dependency: Relying heavily on a handful of dominant upstream AI infrastructure providers (e.g., speci c hyper-scalers or frontier model developers) creates a single point of failure for core business services. Boards should ensure that AI-related risks are incorporated into broader cyber security, operational resilience, and enterprise risk management frameworks.
Boards must oversee a dual challenge: managing massive workforce displacement and morale declines, while simultaneously dealing with an acute shortage of the advanced technical talent required to govern these AI stacks.
2.5.4 Sustainability and ESG headwinds
Dierent AI applications have dierent requirements on computational power, resulting in carbon footprints of dierent sizes. e carbon footprint and water consumption associated with multi-agent architecture choice can directly impact corporate net-zero and ESG commitments. Institutional investors are increasingly scrutinising the exact energy trade-os of corporate AI strategies. Over time, the commercial value of AI applications will become increasingly measurable, predictable and embedded into standard operating models. Boards must be aware of these emerging models and frontier challenges to assess potential disruption or partnership opportunities. Directors do not need deep technical expertise so much as to be able to understand the broad strategic implications. e board’s responsibility is not simply to oversee technology adoption, but to help ensure the organisation remains strategically relevant in an AI-enabled economy.
2.5.3 Workforce disruption
AI has evolved beyond a technology upgrade into a potential driver of workforce disruption and corporate vulnerability. is can create severe organisational friction. • e structural shi in unemployment: Unlike previous automation waves that impacted routine manual labour, frontier AI and agentic workows can disrupt knowledge work, middle management and specialised professional services. • e structural shi in talent for brand management: e proliferation of hyper-realistic deepfakes and advanced disinformation engines means a company’s brand or leadership team can be targeted in highly convincing, automated smear campaigns, impacting stock price before the public relations team can even validate the source.
29
Made with FlippingBook Digital Proposal Creator