# AI Financial Payoff Remains Elusive for S&P 493 Companies Corporate operating models must be fundamentally redesigned before AI investments yield measurable enterprise-wide financial returns. - **Source:** [Fortune | FORTUNE](https://brifing.si/source/769282ef-327a-4454-909a-3c7936216004) - **Published:** 2026-11-06 - **Topic:** [Business](https://brifing.si/c/business) - **Original article:** - **On Brifing:** https://brifing.si/s/245ff850-768e-4e2d-99ad-68e5f99f20ca ## AI Gains Concentrated in Tech Apollo Global Management economist Torsten Slok reports that AI financial benefits are currently limited to the Magnificent 7 tech giants. While these firms see rising revenue per worker, the remaining 493 S&P companies show flat margins. AI spending currently functions as a cost center rather than a driver of profitability for most enterprises. ## The Operating Model Constraint Individual employees are successfully using generative AI to save time, yet these gains fail to scale across organizations. The primary barrier is an outdated operating model built for human pacing. Existing workflows, decision approvals, and handoffs remain unchanged, effectively neutralizing the speed and efficiency advantages that AI technology should provide to businesses. ## CEO Ownership is Essential Successful AI integration requires direct CEO leadership rather than delegation to IT departments. The CEO must act as the accountable party for redesigning work, setting the pace, and determining how much trust to extend to models. Without this high-level authority, organizations lack the mandate to enforce necessary changes across every business function. ## Redesigning Core Business Workflows Companies must deconstruct workflows to eliminate dated processes and redefine roles to share tasks with AI. This involves resetting approval authorities and establishing runtime governance models capable of handling AI output speeds. Leaders must prioritize material workflows that directly impact revenue, margin, risk, or cycle time rather than focusing on interesting technical experiments. ## Infrastructure and Model Strategy Many firms mistakenly use expensive frontier models for routine tasks that could be handled by deterministic automation. Enterprises are increasingly adopting a multi-vendor model strategy, mirroring the shift toward multi-cloud environments. Selecting the right model is not a one-time event, requiring rare organizational capabilities to manage the ongoing mix of open and closed systems. ## The Fallacy of Headcount Cuts Some firms, such as Barclays, have announced job cuts while citing AI, but these layoffs often precede actual productivity gains. Markets may interpret these moves as immediate payoffs, yet the underlying work remains unoptimized. True value comes from outcomes, not just reducing staff, as the build phase currently costs more than it returns. ## Measuring True AI Outcomes Most companies lack the ability to measure the full cost of AI production, which includes data, validation, and human exception handling. Current dashboards track token usage rather than return on investment. Until firms can calculate the cost per outcome, they will continue to struggle with justifying AI spending to their boards and investors. ## Timeline for Widespread Gains Widespread, measurable AI benefits across the S&P 493 are unlikely before the end of 2027. Companies must first master cost-per-outcome metrics, sequence their operating model changes, and enforce model discipline. The flat financial performance observed today will only shift once enterprises move beyond simple adoption and successfully rebuild their core internal processes. ## Sources covering this story - https://fortune.com/2026/11/06/ai-payoff-sp-493-apollo-slok-operating-model-ceos-adam-dumey/ — --- _Summary written by Brifing from the original reporting. Copyright in the original article belongs to its publisher; read the full piece at the link above._