The Dawn of AI’s Impact on Comp Committees

Artificial intelligence isn’t just transforming business models—it’s reshaping the way boards think about executive pay. While no regulations currently mandate linking compensation to AI oversight, the technology’s strategic weight is undeniable. Companies across industries are racing to harness AI, and compensation committees—once focused narrowly on pay and performance—now face an even broader governance role related to talent management and AI. The question is no longer if AI will influence incentive design, but how and how soon.

Oversight Responsibilities

Some boards have explicitly added AI oversight to their committee charters or board agendas. A 2024 survey by ISS-Corporate found that 31% of S&P 500 companies disclosed board committee oversight of AI risks, signaling that governance norms are evolving quickly. In a few cases, some aspect of AI governance falls within the remit of the compensation or talent committee. Comp committees may discuss whether management is appropriately addressing AI opportunities and risks—especially those related to human capital such as reskilling employees for AI use or implementing ethical AI practices to avoid reputational harm.

Incentive Design & Metrics

While standalone AI metrics remain rare, many companies now weave AI objectives into broader performance assessments. Notably, Microsoft’s compensation committee explicitly stated in its 2024 proxy that it “enhanced the FY2025 executive compensation program to align [it] to [the] key strategic priority” of the ongoing AI platform shift.

In practice, Microsoft did not create a separate “AI quota” in the CEO’s bonus formula; instead, the committee adjusted the incentive structure and goals to ensure that success in AI innovation (e.g., deploying generative AI across products) meaningfully impacts pay outcomes. Microsoft’s proxy explained that, to recognize the importance of AI opportunities and challenges, the company wanted a “specific connection of compensation to this key strategic priority.” Microsoft calibrated financial and strategic targets for both annual cash incentives and stock-based awards to consider AI-related growth prospects and execution. If management delivers on AI-driven results, executives are rewarded.

In its 2025 proxy statement, Salesforce disclosed the redesign of its FY26 incentive program to directly link rewards to the strategic execution of Agentforce, its AI agent platform. The CEO’s equity awards are now weighted toward performance targets that include AI-driven “digital labor” transformation.

Most companies—particularly those outside of the tech sector—have not yet adopted explicit AI performance metrics into their incentive plans. Unlike the wave of ESG metrics—for example, diversity or carbon reduction goals that over half of large companies incorporated by 2022—AI metrics remain largely implicit. For instance, many companies use broad innovation or transformation objectives in annual scorecards—AI initiatives often fall under these, without being named “AI” goals. A tech consulting firm might tie a portion of the CEO’s bonus to “digital solutions revenue growth,” which inherently includes AI-powered services, or a bank might include a strategic objective around “technology modernization,” which covers AI deployment in operations. These goals influence pay but aren’t labeled purely as “AI metrics.”

Below is a side-by-side comparison of six non-tech S&P 500 companies showing how AI factored into their 2024 proxy disclosures on executive compensation or strategy, and the nature of that integration.

 

How Boards Factor AI into CEO Evaluations

Even without a formal AI metric, boards frequently discuss management’s progress on AI as part of their qualitative performance assessment. For example, if a CEO successfully integrates AI into product lines or improves efficiency through AI, the compensation committee may exercise its discretion to award a higher annual bonus or an equity grant.

Conversely, if a company falls behind in AI innovation or experiences an AI-related ethics scandal, the committee could reduce payouts or apply a negative modifier for leadership lapses. In 2025, many companies (especially in tech) highlighted AI accomplishments in their business overviews; those achievements often factored into how boards judged the CEO’s strategic leadership for the year. At IBM, for instance, the board’s proxy communications emphasize that IBM’s transformation toward hybrid cloud and AI leadership is a core strategic focus.

While IBM’s incentive plans use traditional financial metrics, the board’s Compensation & Management Resources Committee considers the CEO’s success in advancing IBM’s AI strategy (developing AI offerings and doing so responsibly) as part of the holistic performance review that determines annual pay outcomes. In IBM’s words, “the board is actively engaged in overseeing… [IBM’s] approach to its business, including AI, which we believe must be trustworthy, transparent, and explainable.” This oversight ethos indirectly feeds into how the board sets objectives and evaluates management for compensation; that is, executives are expected to integrate those AI values into business results.

AI-Related Talent and Retention

Compensation committees also find themselves addressing AI from a human capital angle. Increased demand for AI expertise has led some companies to recruit high-priced AI specialists and upskill their workforce. OpenAI provides one example. The company’s equity-based pay is exceptionally high, averaging approximately $1.5 million per person for its 4,000 employees, according to a recent Wall Street Journal/Equilar comparison. This figure is roughly 34 times the average seen at 18 other major tech firms during their pre-IPO phases over the last quarter-century, the WSJ reports.

Additionally, existing executives taking on major AI initiatives may receive retention bonuses or higher pay to reflect their new responsibilities. For example, if the CTO is now also driving an enterprise AI transformation, the committee might increase that CTO’s long-term incentive target to keep them on board in a hot talent market. These actions aren’t always detailed in proxy materials, but they represent a direct way in which AI is shaping committee decisions on pay.

AI’s Ripple Effect on Workforce Incentives

Many compensation committees have, in recent years, expanded their oversight beyond the C-suite to the wider workforce, often renaming themselves to “Compensation and Talent Committee.” In this context, committees are considering how AI will impact jobs, skills, and incentive structures at all levels. The rapid adoption of generative AI in business processes presents a double-edged sword: efficiency gains but also employee anxiety about job security and skills.

Proactive committees overseeing company-wide pay and benefits are beginning to discuss how to incentivize reskilling in AI and ensure that performance metrics don’t inadvertently discourage prudent AI use. For instance, a sales team might use an AI tool to generate leads; a committee overseeing sales compensation plans will want to ensure the plan rewards effective integration of such tools, not just raw sales, so that employees aren’t penalized for spending time learning new AI systems.

According to KPMG’s Board Leadership Center, in 2025 compensation committees should ask whether management has strategies to address “the impact that GenAI and other emerging technologies will have on the company’s workforce, including recruitment and retention strategies for necessary technological expertise [and] employee concerns about job elimination.” Such oversight falls squarely under the committee’s responsibilities for human capital oversight. Some comp committees are also reviewing whether incentive goals remain appropriate in an AI-augmented environment (e.g., if AI automation dramatically boosts productivity, should performance targets be raised?) The overall pay philosophy may shift to emphasize innovation and adaptability. In short, comp committees are thinking not only “Do we pay competitively to AI experts?” but also “Are we creating the right incentives for our entire workforce to embrace AI-driven change?”

Risk Management and Clawbacks

Another angle is ensuring that compensation does not encourage irresponsible AI behavior. Just as boards don’t want pay plans that encourage excessive risk-taking in finance, they also don’t want to inadvertently push executives to deploy unproven AI recklessly to hit a short-term goal. With the Securities and Exchange Commission’s (SEC) new clawback rules and long-standing malus provisions, compensation committees have tools to recoup pay in the event of a major compliance or ethical failure. If, say, a company’s AI product caused a legal or reputational crisis, the compensation committee could exercise discretion to reduce payouts.

Glass Lewis’s 2024 policy update on clawbacks suggests boards should have the power to cancel or recoup pay for “evidence of problematic decisions or actions,” including material risk management failures. One could envision the inclusion of AI-related oversight failures. While hypothetical, the ability to claw back or rescind pay for such failures underscores the role of the compensation committee as part of the essential governance checks and balances around AI.

Regulatory Guidance and Investor Views

As of 2025, there are no specific SEC mandates linking AI oversight to executive compensation. The SEC’s focus on AI has been around disclosure (e.g., urging transparency of AI risks and opportunities in filings) and warning against “AI hype.” That said, general principles of disclosure and risk management apply. If AI materially changes a company’s risk profile or strategy, boards should disclose how they’re overseeing it, including how they incentivise management to address it. For example, if AI is central to a company’s future, the SEC would expect discussion in the 10-K Management Discussion & Analysis (MD&A) or proxy about that strategy; part of that might include whether executive pay programs are aligned to executing the AI strategy.

In Microsoft’s 10-K, the company touted massive AI investments and launches, and in the proxy, it coherently showed how the board linked those to CEO incentives. This consistency between narrative and pay design is viewed positively by regulators and investors.

Proxy advisers and institutional investors are also signaling expectations. Glass Lewis, in late 2023, updated its voting guidelines to note that boards should be equipped to oversee emerging risks like AI. The proxy advisor flagged that it would monitor failures of oversight, which could include mismanagement of AI, as possible reasons to recommend against directors (i.e., potentially members of relevant committees). While Glass Lewis did not say “tie CEO pay to AI,” it expects boards to manage AI risk. Investors might question a company with significant AI exposure if there is no mention of board oversight or alignment in strategy.

So far, shareholders have not introduced proposals seeking AI-linked pay metrics, unlike in climate governance, where some investors asked to tie executive compensation to emissions targets. The AI-related shareholder proposals in 2023–2024 instead sought governance transparency (e.g., reports on AI ethics or risks). For example, at Apple’s 2024 meeting, investors proposed a report on AI governance. Apple’s board opposed it, explaining that its existing risk oversight by the full board covers AI issues. The proposal received about 40% support, a sign of substantial investor interest. Notably, Apple’s response did not mention executive compensation; it focused on board oversight through committees, suggesting that, at least for now, investors are pushing for oversight structures and disclosure on AI rather than an explicit pay-for-AI-performance linkage. Nonetheless, by establishing that AI is overseen at the highest levels, boards implicitly assure shareholders that management will be held accountable should they fail to navigate AI properly. Such accountability could include consequences for what’s been paid (clawbacks) or what is promised to be paid (malus).

Looking ahead, more explicit connections are to be expected. If AI becomes a key driver of long-term corporate value in an industry, compensation committees may introduce related metrics, akin to how some companies added cybersecurity objectives after high-profile breaches, or how ESG metrics were integrated when sustainability rose on the board’s agenda.

As board oversight mechanisms mature, they could evolve into concrete features of an incentive plan. For instance, a company might set a goal for “AI-driven revenue growth” or “the percentage of products with embedded AI” as a formal bonus criterion once they have a baseline to measure against. Any such metric would need to be carefully designed to drive the desired behavior and discourage recklessness.

Actionable Takeaways

• Reward responsible AI adoption: Ensure incentive plans recognize executives who successfully integrate AI into strategy—while maintaining ethical and risk-aware practices
• Signal alignment in disclosures: Follow best practices like Microsoft by clearly linking AI priorities to pay programs in proxy statements for transparency and investor confidence
• Secure AI talent strategically: Approve competitive packages for critical AI roles and retention bonuses for executives leading major AI initiatives—while balancing cost and governance
• Future-proof pay philosophy: Reassess performance metrics and incentive structures to reflect AI-driven productivity gains and innovation, avoiding outdated targets
• Address workforce impact: Incorporate reskilling incentives and review company-wide pay plans to encourage adoption of AI tools without penalizing learning time
• Embed risk controls: Use clawbacks and malus provisions to deter reckless AI deployment and protect against compliance or reputational failures
• Stay ahead of investor expectations: Monitor evolving shareholder and proxy advisor guidance on AI oversight and be prepared to demonstrate accountability in both governance and compensation