A director fed five years of financial data into ChatGPT and spotted a pattern management’s reports hadn’t highlighted. Her excitement quickly turned to uncertainty: was she overstepping?
Korn Ferry recently explored this scenario, and it gets at something real. Every AI-aware director now faces this tension. You have a fiduciary duty to ask probing questions using the tools available to you. But you also need to maintain the trust-based relationship with management that makes boards work. Use AI to dig too deep, and you look like you’re second-guessing the team. Don’t use it at all, and you might miss something you had a duty to catch.
The Dilemma Is Real
According to Korn Ferry, 40% of leaders are already using AI in their daily work. Directors have discovered they can summarize materials, benchmark performance, and surface patterns in data faster than ever. But this newfound capability raises hard questions about governance boundaries.
The business judgment rule has long protected directors who rely on information management provides. But when directors can easily access deeper analysis, do they have an obligation to look beyond curated materials? As Korn Ferry’s Anthony Goodman frames it: “Should they use it? What will be the legal jeopardy if they do? Or don’t?”
The Risk Goes Both Ways
Too far: Directors who run their own alternative financial models or conduct independent analysis may undermine management’s confidence. Teams that already feel directors get stuck in the weeds will become less transparent if they sense their work is being second-guessed. That erosion of open information flow is itself a governance failure.
Not far enough: Directors who avoid AI entirely may fail to meet evolving standards of fiduciary duty. If the tools exist to catch a risk earlier, and a director chose not to use them, that’s a harder position to defend - especially after something goes wrong.
The balance, as Korn Ferry’s Jamen Graves puts it: “The key is to strike a balance between moving forward the agenda set by management and the board’s interest in ensuring a broader external perspective.”
The Security Problem Makes It Worse
There’s a complication most articles about “AI in the boardroom” ignore: security. Directors using ChatGPT or other consumer AI tools with confidential board materials risk exposing that information to training data, account compromise, and compliance violations.
Board materials are among the most sensitive documents a company produces - strategic plans, M&A discussions, executive compensation, financial projections. Pasting them into a public AI tool is a governance failure in itself, regardless of the analytical benefit.
Most directors don’t have access to enterprise AI systems configured for sensitive information. So they face an impossible choice: use consumer tools and accept the security risk, or avoid AI entirely and accept the preparation gap.
Understanding, Not Analyzing
This is where the distinction matters most. There’s a critical difference between:
- A director who uses tools to better understand management’s financial presentation vs. one who runs their own alternative financial models
- A director who asks informed questions based on deep preparation vs. one who challenges management’s methodology
- A director who identifies strategic blind spots for board discussion vs. one who conducts independent strategic analysis
The first version in each pair makes directors more effective within their governance role. The second pulls them across the line into management’s domain.
The right approach helps directors deeply understand what management has prepared - surfacing the right areas for discussion, identifying where questions would be most valuable, and connecting current materials to past decisions and external context. It doesn’t hand directors tools to run parallel analysis.
What This Looks Like in Practice
Before the meeting: You review a quarterly financial presentation with AI-generated context. The AI highlights what changed significantly, connects metrics to trends from previous quarters, and flags areas where the board might want deeper discussion. You arrive with informed questions, not independent conclusions.
During the meeting: Your questions are sharper because you understood the materials more deeply. You ask about the implications of a revenue trend, not about what drove the variance. Management recognizes the quality of engagement and provides more candid answers. Trust strengthens.
Between meetings: You maintain awareness of developments relevant to your committee role without needing to request special briefings. When an urgent issue arises, you can engage quickly with appropriate context.
None of this requires you to step outside your governance role. You’re not conducting independent research or second-guessing management’s analysis. You’re understanding it well enough to do your job - which is oversight, not operations.
The Path Forward
The AI director dilemma isn’t going away. But the solution isn’t to choose between being well-prepared and maintaining trust. It’s to use tools specifically designed for this middle ground - tools that enhance understanding without encouraging overreach, that provide security appropriate for the most sensitive information, and that respect the boundary between governance and management.
Directors who find this middle ground fulfill their fiduciary duties more effectively while strengthening the collaborative relationship that makes boards valuable. That’s not a compromise. It’s what good governance looks like in an AI-enabled world.
Aureclar was built for the middle ground - understanding, not analyzing. See how it works