Key Takeaways
- The top 1 % of U.S. firms spend a median of $7,400 per employee on AI, more than 600× the typical business’s $11.95 per employee.
- Anthropic dominates AI spending, adopted by 43.5 % of U.S. businesses in July, up 1.1 pp month‑over‑month.
- Even the biggest spenders hit a price ceiling: Anthropic’s premium model Fable 5 ($10 / M tokens) captured only 6 % of token purchases and 11.4 % of dollars spent on Anthropic models one month after launch.
- Cheaper, open‑source and Chinese‑developed models are slowly gaining traction, now used by 6.1 % of AI‑adopting firms.
- CFO optimism is rising: 39.1 % now expect very positive generative‑AI returns within 1‑2 years (up from 0 % in mid‑2025).
- Organizational readiness, not the technology itself, is cited by 71 % of senior tech executives at $1 B+ firms as the main barrier to AI performance.
- Only 2 % of S&P 500 companies quantified AI effects in Q2 earnings; of those, 11 % reported measurable productivity gains in coding or support.
- Early adopters that do measure AI see median earnings rise 17 %, modestly ahead of the 14 % gain for non‑measurers.
- The AI landscape is bifurcated: a small cohort bets heavily while most firms purchase cautiously.
Overview of AI Spending Disparities
The Ramp AI Index released on August 12 reveals a stark divide in how much U.S. companies are willing to invest in artificial intelligence. “The top 1% of U.S. businesses spent a median of $7,400 per employee on artificial intelligence in July, while the top 10% spent roughly $650,” the report states. By contrast, the median company spent just $11.95 per employee. This yields a per‑employee spending gap of more than 600 × between the heaviest spenders and the average firm, and roughly 11 × even when compared with the next tier of aggressive adopters. Such disparity underscores a market where a minority of firms are making sizable, targeted bets on AI while the majority remain cautious, testing the waters with minimal outlays.
Anthropic’s Market Share and Growth
Anthropic has emerged as the clear leader in business AI adoption. According to Ramp, “Anthropic is capturing the biggest share of that money. It leads overall business adoption, with 43.5% of U.S. businesses paying for its subscriptions or tokens as of July, up 1.1 percentage points from the prior month.” This steady climb reflects growing confidence in Anthropic’s model offerings and its ability to capture enterprise budgets. The firm’s share now outpaces rivals by a considerable margin, suggesting that its go‑to‑market strategy—emphasizing reliability and enterprise‑grade support—is resonating with decision‑makers looking for proven AI partners.
OpenAI and xAI Trends
While Anthropic leads, OpenAI continues to hold a significant presence, albeit with slower growth. The data show that OpenAI “trails after adding just 0.23 percentage points over the same period,” indicating a more modest uptake compared with Anthropic’s surge. Meanwhile, xAI, the newer entrant backed by Elon Musk, “notched its fastest growth since July 2025, reaching 4% of businesses.” This rapid ascent, though from a low base, signals that alternative providers are beginning to carve out niches, especially among firms seeking differentiated capabilities or pricing models that diverge from the established duopoly.
Price Sensitivity and Model Adoption Ceiling
Even among the biggest spenders, there is a clear ceiling on how much businesses will pay for the latest, most capable models. Anthropic unveiled its flagship Fable 5 in July at roughly $10 per million tokens, twice the price of OpenAI’s GPT‑5.6 Sol. One month after launch, “Fable 5 made up only 6% of the tokens businesses purchased from Anthropic and 11.4% of total dollars spent on Anthropic models,” Ramp found. By contrast, OpenAI’s lower‑priced GPT‑5.6 Sol “captured 25% of OpenAI’s tokens and 23% of its spend over the same period.” As Ramp’s lead economist Ara Kharazian put it, “So with Fable 5, we’ve found a new upper bound for how much businesses are willing to spend on AI… the labs will need to prove performance beyond what even Fable 5 is able to achieve and simultaneously ensure that competitors aren’t able to come reasonably close.” This commentary highlights that price performance trade‑offs are now a decisive factor in model selection.
Shift Toward Cheaper Alternatives
Despite the premium pricing of leading models, a quiet migration toward more affordable options is underway. Adoption of model‑serving platforms that offer open‑source and Chinese‑developed models “kept climbing, reaching 6.1% of AI‑adopting businesses, up 0.2 percentage points from the prior month.” While still a small fraction, the steady uptick suggests that cost‑conscious enterprises are experimenting with alternatives that can deliver adequate performance at a fraction of the expense. This trend may accelerate if the performance gap narrows or if enterprise procurement policies tighten further.
Evolving CFO Expectations on ROI
Financial leaders are becoming more optimistic about the payoff from AI investments. “The share of CFOs expecting very positive returns from generative AI within one to two years jumped from zero to 39.1% since mid‑2025,” according to PYMNTS Intelligence, while the share anticipating a three‑to‑five‑year horizon fell from 65.9% to 34.8%. This shift indicates that early successes—particularly in automation and productivity gains—are reshaping the narrative from a long‑term experiment to a near‑term value driver. Nonetheless, the majority still view returns as a multi‑year endeavor, reflecting the inherent complexity of integrating AI at scale.
Organizational Readiness as Bottleneck
When asked what limits AI performance, senior technology leaders point inward rather than to the technology itself. PYMNTS Intelligence’ Enterprise AI Benchmark Report found that “71% of senior technology executives at companies with at least $1 billion in annual revenue say organizational readiness, not the AI itself, is the primary factor limiting performance.” Only 11% blamed the technology, suggesting that most firms see the challenge as one of change management, data governance, skill development, and process redesign. Addressing these internal hurdles appears to be a prerequisite for unlocking the full potential of AI investments.
Early Evidence of AI Impact on Earnings
Quantifiable financial benefits are beginning to surface, though adoption of rigorous measurement remains limited. “Only 2% of S&P 500 companies quantified the effects of AI in their second‑quarter earnings reports, and of those, 11% cited measurable productivity gains in areas such as software coding or customer support,” PYMNTS reported, citing a Goldman Sachs analysis. Among the firms that do measure, “median earnings rose 17% at companies that quantified again, compared with 14% for those that did not.” The modest gap hints that early movers who institute accounting discipline for AI are already pulling ahead, even before most peers have built the capability to prove ROI systematically.
The Two‑Speed AI Landscape
Taken together, the data paint a picture of an AI market operating at two velocities. A small cohort of firms is allocating substantial resources—spending thousands per employee on premium models and aggressively testing cutting‑edge capabilities—while the majority procures AI cautiously, favoring lower‑cost alternatives and limiting spend to modest per‑employee amounts. This bifurcation creates a competitive dynamic where early, well‑funded adopters may secure first‑mover advantages in efficiency, product innovation, and market share, whereas slower movers risk falling behind if they cannot close the readiness and investment gaps.
Outlook and Implications
Looking forward, several forces will shape the trajectory of business AI. Continued pressure on model providers to demonstrate clear, price‑justified performance gains will likely keep premium adoption constrained unless breakthroughs emerge. Simultaneously, the steady rise of open‑source and regionally specific models suggests a growing democratization of AI access, potentially leveling the playing field for cost‑sensitive enterprises. For AI to deliver on its promise, companies will need to pair technology investments with concerted efforts to upgrade organizational readiness—upskilling workers, refining data pipelines, and aligning incentives. Those that succeed in marrying spending with structural change are poised to capture the outsized returns that early metrics already hint at.