May 13, 2025
Managing Partner
Boards that actively steer generative-AI adoption, pairing disciplined value creation with uncompromising risk oversight, will define their companies' competitive trajectory.
Boards today sit at the confluence of an extraordinary opportunity and an equally formidable set of risks. Generative AI's breakthroughs, a developer who prompts a model to draft thousands of lines of production-ready code, a litigator who unearths precedent in minutes, a biochemist who generates a novel protein sequence at the click of a button, leave even seasoned executives awestruck. Yet the same algorithms can hallucinate falsehoods, replicate hidden bias, or expose proprietary data to prying eyes. As the guardians of long-term enterprise value, directors must help management harness the upside while avoiding the pitfalls. Doing so requires a structured line of inquiry and a willingness to deepen their fluency with the technology.
Below, we expand the questions directors should pose to management, re-examining them with the same depth and analytical rigor that Santiago & Company brings to any strategic issue. We conclude with a parallel challenge for the board itself. The result is a roadmap rich enough to guide debate in the next quarterly meeting yet practical enough to spur action on Monday morning.
Directors cannot delegate strategic foresight. They must insist that management quantify both the short-term efficiency gains and the longer-term disruptions that generative AI will unleash across the value chain.
Santiago & Company's latest models suggest that software engineering, marketing and sales, customer service, and product development will feel the first wave of impact. A global bank that rewrites legacy code with AI-assisted tools already reports development cycles 40 percent faster. A consumer-electronics brand now tests hundreds of promotional variants in a single afternoon, translating copy into dozens of languages with near-human nuance. Even sectors once thought insulated, heavy manufacturing, utilities, and logistics, see AI copilots drafting work instructions, summarizing maintenance logs, or optimizing shipping routes. Momentum builds quickly: ChatGPT reached 100 million users in eight weeks, dwarfing the adoption curves of Instagram, TikTok, and previous platform phenomena.
As models scale and specialized datasets proliferate, directors should expect competitive boundaries to blur. Media libraries become training corpora; proprietary sensor readings feed predictive models; entire operating platforms integrate generative capabilities. In energy, for instance, field data from wind turbines now inform design suggestions for the next-generation blade. In life sciences, multimodal models rapidly fuse text, imagery, and biochemical structures to generate novel molecular candidates. Boards must press management to chart scenarios: What happens when an insurgent rivals our end-to-end digital experience? Where does our current advantage withstand algorithmic commoditization, and where does it erode?
Generative AI's risk profile widens well beyond the concerns that accompanied traditional machine learning. Directors should insist on dual lenses: a rigorous value-at-stake assessment for each use case and an equally rigorous catalog of downside exposures.
Hallucination and misinformation. Large language models can "confidently" fabricate citations or numeric facts. A pharmaceutical company chatbot that discloses an inaccurate dosage or a financial services tool that misquotes regulatory thresholds could trigger severe liability. Boards must ensure mechanisms exist to detect hallucinations early, quarantine suspect outputs, and surround algorithms with human review.
Intellectual property infringement. Training data often scrape copyrighted materials in ways that courts have yet to fully adjudicate. An enterprise model that ingests trade publications might inadvertently reproduce protected phrasing. Directors should confirm that legal teams have inventoried training sources, captured vendor indemnification agreements, and established governance for future data ingestion.
Cybersecurity and deepfake vectors. Generative AI reduces the cost of phishers' creativity. Synthetic audio that mimics a CFO's cadence can direct real-time wire transfers; manipulated video can crash consumer trust overnight. Boards should mandate "red-team" exercises that stress-test protocols against AI-enabled threats.
Environmental impact. Scaling a model from a billion to a trillion parameters multiplies compute requirements and carbon emissions. Sustainability pledges must encompass AI workload optimization and renewable-energy commitments for data-center partners.
A robust posture pairs value creation with clearly articulated guardrails. Management should present tiered approval pathways for high-, medium-, and low-risk applications, codified testing protocols, and a playbook for rapid suspension should an unforeseen risk surface.
Many companies once approached AI through scattershot pilots, which were useful for experimentation but inadequate for enterprise transformation. Generative AI's velocity and systemic reach demand a different architecture.
Executive ownership. Boards should appoint or confirm a single senior leader, chief AI officer, chief technology officer, or business-unit head empowered with cross-functional authority to govern generative AI strategy. This individual must control the budget, set priorities, and resolve risk-return trade-offs.
Cross-functional alliance. A foundation model rarely remains confined to an R&D lab. Legal, compliance, cybersecurity, HR, brand, and line-of-business leaders must convene weekly, if not daily, to steer a shared backlog of use cases. Centralized coordination prevents the "shadow model" creep that plagued earlier AI efforts.
Partnership strategy. The generative ecosystem shifts monthly. Specialized model vendors emerge, cloud hyperscalers release new accelerators, and start-ups bundle domain-specific data and tuning services. Boards should probe for an alliance map that protects flexibility, modular interfaces, and open-standard APIs to avoid lock-in while accelerating deployment. Vendor-risk assessments must extend to the training-data supply chain, inference pipelines, and downstream outputs.
Generative AI feeds on data liquidity. Legacy enterprises still wrestle with fragmented warehouses, redundant ingestion pipelines, and inconsistent data taxonomies. Directors ought to demand clear milestones for:
The skills equation is neither linear nor static. A single transformative use case may require prompt engineers, domain scientists, machine-learning architects, and product-design specialists. Simultaneously, existing roles from financial analysts to customer agents will absorb AI copilots that augment human judgment.
Directors should ask:
Generative AI flourishes in environments that reward experimentation, tolerate fast-cycle failure, and anchor decisions in data. Suppose the current culture tilts toward hierarchy or risk aversion. In that case, boards may need to champion new norms of lightweight governance for pilot launches, transparency about model performance, and open forums where employees challenge algorithmic outputs without repercussion. Ethical stewardship must become everyone's responsibility, not merely a compliance afterthought.
Even directors who oversaw prior digital transformations acknowledge that generative AI represents a paradigm shift. To govern effectively, boards must elevate their capabilities through three mutually reinforcing actions.
Refresh composition. Traditional board matrices, which include industry experience, financial acumen, and a global perspective, now require an additional column: fluency in advanced analytics and machine learning. Recruiting a director steeped in deep-tech ventures or academic research can accelerate collective understanding. Where such expertise remains scarce, advisory councils offer a practical bridge.
Commit to ongoing education. Santiago & Company often facilitates immersive workshops that dissect a live model's architecture, demonstrate prompt engineering, and examine real-world failure modes. Directors who experience the tools firsthand, drafting agenda briefings and synthesizing 200-page strategy decks into executive summaries, gain intuition that no slide presentation can replicate.
Incorporate AI into board workflows. The boardroom provides an ideal sandbox. A generative AI assistant might surface contrarian arguments for a capital-allocation decision, summarize competitive intelligence overnight, or test the logic of a proposed acquisition against historical analogs. Hands-on use builds empathy for management's challenges and grounds oversight in practical knowledge.
Generative AI is advancing at a pace that compresses strategic planning cycles from years to quarters, sometimes weeks. Directors need not master every technical nuance, yet they must foster a governance environment where curiosity outruns complacency and rigor checks exuberance. By interrogating strategy, risk, organization, capability, and readiness with depth and persistence, boards can steer their enterprises toward the technology's vast upside while shielding stakeholders from its darker edges.
The coming months will separate companies that embed generative AI at the core of their business model from those that dabble at the margins. Boards that embrace their responsibility, challenge management constructively, guide resource allocation, strengthen talent, and model ethical leadership will amplify value creation for shareholders and society. Santiago & Company's experience across sectors affirms that disciplined questions today lay the groundwork for sustainable advantage tomorrow.
Generative AI is rewriting the rules of corporate strategy and boards that hesitate to take risks. This article reveals how directors can capture the technology's multitrillion-dollar upside while shielding their organizations from its gravest threats.
Generative AI is rewriting the retail playbook: brands that personalize every touchpoint are pulling away from the pack. Discover how Santiago & Company’s latest insights reveal the moves that turn AI investment into measurable growth and lasting customer loyalty.
Scaling generative AI demands CIOs shed low-value pilots, streamline technology and data practices, and embed responsibility at every step to unlock real business impact.
Heightened trade disputes and global uncertainties demand more than short-term fixes and reactive risk measures. Organizations that embed resilience throughout their strategy, operations, and culture will thrive amid ongoing volatility and position themselves for enduring success.
