Boards say the C-suite owns the AI strategy. The C-suite doesn’t agree

A Pearl Meyer survey of 108 executives and board members found 90% of directors believe AI ownership belongs with the C-suite, but executives are split on accountability: 32% say the C-suite owns AI strategy, 27% point to individual business leaders, and 17% to functional heads. The survey also shows a governance gap, with 100% of directors calling their senior team cohesive versus only 66% of executives, and just 54% saying priorities clearly reach leaders two levels below the C-suite. Overall, 40% of companies are still piloting AI and 31% are using it ad hoc, suggesting deployment is being slowed by internal coordination rather than the technology itself.

Analysis

The market is still pricing AI as a top-line revenue and margin lever, but this piece highlights a more important bottleneck: organizational throughput. When ownership is diffuse, the near-term winners are not the obvious AI adopters but vendors that monetize governance friction—security, data plumbing, workflow orchestration, auditability, and model controls. That favors picks-and-shovels software more than pure “AI story” names, because enterprise rollout delays stretch the monetization window while increasing spend on readiness, not just inference.

Second-order, the biggest risk is not that AI fails; it’s that companies use AI as a post hoc explanation for restructuring and miss actual productivity measurement. Over the next 6-18 months, that creates a gap between headline AI announcements and realized KPI impact, which can compress multiples for firms trading on “AI efficiency” narratives if the reported savings don’t show up in SG&A, headcount, or cycle-time data. META and ORCL have some support from AI-driven capex and efficiency optionality, but the bar rises quickly if internal governance remains fragmented and the market starts demanding proof rather than intent.

The contrarian read is that this is not a bearish AI signal, but a timing signal: adoption is likely slower than consensus, yet more durable once scaled because the hurdle is managerial, not technical. That means the next few quarters should reward companies with centralized decision rights and strong data infrastructure, while penalizing firms that treat AI as an informal productivity push. In practice, the dispersion should widen between operators that can instrument ROI and those that can only narrate it.

For XYZ, the neutral data profile suggests no company-specific edge here; the actionable angle is relative exposure to enterprise software and infrastructure names that sell governance, security, and workflow control into cautious buyers. The key catalyst is earnings season: any surprise in AI-related spend, implementation timelines, or productivity disclosures will likely move these stocks more than generic AI commentary.

Key Decisions for Investors

• Long ORCL vs. a basket of high-multiple AI application names for 3-6 months: ORCL should benefit if enterprise AI spend shifts toward infrastructure, security, and controlled deployment; risk/reward improves if guidance emphasizes backlog conversion rather than headline AI usage.
• Add to META only on evidence of measurable AI-driven margin expansion, not narrative alone: use a staggered entry over the next 1-2 earnings cycles, with a tight stop if operating expense deleverage reappears or AI efficiency claims remain unquantified.
• Buy cybersecurity/data-governance beneficiaries on weakness over the next 1-3 months: favor names exposed to model controls, data quality, and audit trails; these are the first-order spend items when companies move from pilots to rollout.