Board Leadership in the AI Era
Organizations Must Evolve Third-Party Risk Oversight as Vendors Rapidly Embed AI
Artificial intelligence has reached a critical inflection point where board-level engagement is no longer optional—it’s a fiduciary responsibility. As AI fundamentally disrupts competitive landscapes, operational models, and strategic positioning, boards must evolve from passive overseers to active architects of their organizations’ AI transformation. The companies that will thrive are those whose boards recognize that inaction itself has become a risk that can lead to disruption.
The Strategic Imperative for Board Engagement
The disruptive power of AI is now undeniable, with companies actively seeking to transform products, services, operations, and workflows through this technology. Recent data shows that 32% of global CEOs report generative AI has already increased revenue, while 34% cite increased profits.
CEOs report generative AI has increased revenue
CEOs cite generative AI has increased profit
Yet this same technology that promises transformation also introduces unprecedented risks—from employees inadvertently uploading confidential material into AI systems to the emergence of sophisticated security threats like prompt injection attacks.
The velocity of AI adoption has caught many boards unprepared. One stark example emerged when an online education company saw its share price cut in half in a single day after warning that ChatGPT would impact subscriber growth as students turned to generative AI for homework assistance. This dramatic market reaction underscores how AI can rapidly reshape entire business models and competitive dynamics, making board oversight not just prudent but essential for protecting shareholder value.
Boards face a crucial responsibility in overseeing the key risks associated with AI deployment while ensuring management implements appropriate controls. As custodians of enterprise risk management, directors must be prepared for the potential dangers while fostering a spirit of experimentation and exploration that enables value creation.
Six Critical Areas for Board Focus
1
The full board should typically oversee AI business transformation given its strategic and risk implications, though specific aspects may be delegated to committees—audit for enterprise risk, technology for tools and products. Organizations must establish clear management accountability with defined board touchpoints. Leading boards are adding AI implementation to CEO objectives and key results, creating direct accountability for achieving transformation goals.
This requires boards to continuously assess and develop their own AI knowledge. Directors need sufficient understanding to ask penetrating questions about strategic direction, challenge management assumptions, and recognize when the organization may be falling behind competitors or taking undue risks.
2
AI's transformational power lies not just in productivity gains but in unlocking new growth and business strategies. Boards must understand management's process for identifying and prioritizing AI opportunities, oversee strategic assessment and planning, and evaluate resource investment levels. Critical questions that haven't been asked in years must now be on the table: Who are the emerging competitors? What kind of company will we become? How do we serve customers differently? What ecosystem partnerships do we need?
When generative AI enables customers to request services through chat prompts that can also write execution code, fundamental business model questions arise. Will a hotel chain become a comprehensive travel portal? Will it compete directly with e-commerce players? These strategic pivots require board-level consideration of not just AI capabilities but entire partnership ecosystems and business plans.
3
All AI models are based on data, making proprietary first-party data—especially about customers and their behaviors—pure gold. This data enables better customer prioritization, trigger identification for follow-up, and personalization of interactions. The more organizations interact with and capture information about customers, integrating data across the enterprise and creating new data through testing, the richer their AI models become.
Boards must demand deep examination of data management processes: how data is captured or purchased, what happens to it, how it's enhanced, maintained, secured, and activated. Leading organizations are adapting business models to create and capture more granular information faster on more customers and operations. They're analyzing what data needs permissioning and how permissions are managed, while addressing security and privacy concerns as data flows through growing lists of AI models.
4
Adopting responsible AI practices is critical for enabling consistent, well-informed decisions that allow organizations to execute AI strategies at speed. Boards must ensure there's a proper risk foundation, diving deeper on AI models that create higher risk, and understanding how management fosters stakeholder trust.
This includes addressing evolving regulatory landscapes, from emerging mandates to contract and legal implications. Without clarity on how AI models have been trained, users face legal risks from inadvertently using copyrighted material. While watermarking techniques are emerging and some technology players offer indemnity, the law remains ambiguous, leaving many organizations exposed to cease-and-desist orders from original content creators.
Key governance considerations include requiring transparency on training sources, regular bias assessments, alignment with privacy standards like the Children's Online Privacy Protection Act, and guarantees that employees won't train other companies' models on proprietary data or enter first-party information into open systems.
5
AI's potential to completely transform traditional workforce models demands proactive board engagement. The hardest part of any organizational transformation is culture change, and AI transformation is no exception. Success requires focusing on new skills, talent sourcing, development, measurement, and significant cultural transformation.
Technology companies are already seeing reduced need for product managers while requiring more data engineering and data science talent. Boards must engage leadership on AI-related skills and competencies needed, reskilling and upskilling approaches for existing workforce, practical talent sourcing and acquisition strategies, and adoption of AI-related ethics, privacy, and security policies across the organization.
Critical to this transformation is ensuring communications happen often and early, understanding how talent strategies must change, and discussing approaches to productivity gains. The supply-demand gap for AI talent will only intensify, making proactive talent and compensation planning essential.
6
Performance monitoring is crucial to verify investments yield desired outcomes. While companies may not yet have fully established systems for monitoring AI success, boards must prevent AI initiatives from becoming financial black holes. Regular reporting on progress and metrics to evaluate success are essential.
Leading organizations are creating cross-business pools for forward-moving AI investments rather than making incremental changes to funding across product lines. This might include stitching sensors into equipment, building cloud infrastructure for client data, or establishing innovation funds for exploring new tools. AI has become a standing agenda item at board meetings, with conversations focusing on implementation speed and progress monitoring.
Shaping the Industry Ecosystem
Integrating AI immediately raises strategic questions about what to build versus buy. Most companies cannot build their own large language models and will need to tap major technology vendors for chips, systems capacity, training data access, and AI development talent. Focused applications—from specialized content generation to chatbot functionality—will come from a mix of off-the-shelf applications, custom developers, and in-house teams.
Organizations are testing offerings from major LLM players who increasingly offer openly trained foundations as a base, upon which users can further teach and optimize systems. These vendors provide partitioned model instances, protecting company data uploaded from proprietary systems. However, boards must recognize the many conditions requiring consideration in these partnerships.
Is your board equipped to govern in the age of AI?
Discover essential frameworks for strategic oversight, risk management, and AI governance at the highest level.