Ramp Report Shows AI Spending Flatlining Despite Rising Token Volumes

New corporate spending data indicates that artificial intelligence usage continues to scale upwards, yet financial investments in the technology have flatlined [1]. According to a report published by Ramp, token spending grew by just 1.5% across August and September, representing a sharp deceleration from the 42% month-over-month growth recorded in July [1].

Shifting Market Dynamics and Price Wars

The recent data suggests that aggressive price competition and savvier customer cost controls are successfully capping revenue growth for major artificial intelligence developers like OpenAI and Anthropic [1]. Ramp lead economist Ara Kharazian noted that the era of hypergrowth may be concluding because the top tier of heavy spenders has encountered a ceiling [1]. While October data indicated a minor re-acceleration with token spend up 8.9% between September 23 and October 23, broader headwinds constraining spend growth remain robust [1].

The Impact of Thrifty Usage and Cheaper Models

The spending slowdown is not driven by a lack of customer demand, as token volumes continue to rise [1]. Instead, market dynamics have shifted toward a buyer's market [1]. OpenAI slashed API prices in half for models such as GPT-6 Sol and GPT-6 Luna, while Anthropic reduced pricing for Fable 5.1 and Opus 5.5—a move Anthropic stated in a regulatory filing cut its revenue growth by approximately $600 million [1].

Compounding these cuts, enterprise customers are deploying sophisticated cost controls [1]. Organisations are implementing automated routing to direct standard tasks to cheaper light models, setting employee budgets, and utilizing default settings that restrict usage of expensive models for long-running processes [1]. Additionally, alternative tools such as lightweight routing frameworks like Jev are capturing top performance metrics by helping companies handle basic workloads at minimal cost [1].

Investor Re-evaluations and Valuation Pressures

These cost-efficiency trends coincide with downward pressures on major AI firm valuations and fundraising targets [1]. Bloomberg reported that Anthropic's potential valuation declined from an early October estimate of $2 trillion down to $1.2 to $1.3 trillion, alongside expectations of receiving a high junk rating from credit rating agencies [1]. Meanwhile, OpenAI lowered its proposed valuation to $1.25 trillion from $1.4 trillion and reduced its recent fundraising goals by approximately $10 billion [1].

The corporate adoption of artificial intelligence has expanded rapidly over recent years, driven largely by venture-backed, high-growth technology companies utilizing advanced coding agents and language models [1]. Major model providers have historically subsidized high-end technology use and offered enterprise discounts to secure customer retention [1]. However, intense competition from open-source models, alternative tech giants including Meta and Nvidia, and a proliferation of domestic and international open-model providers have transformed the landscape into a buyer's market [1].

For industry observers and market participants in India and globally, the decoupling of token volume growth from revenue growth challenges fundamental assumptions regarding the hypergrowth trajectory of foundational AI companies [1]. As major developers approach initial public offerings and capital raising milestones, tighter corporate spending controls and widespread price discounting directly impact revenue projections, valuation metrics, and broader sector economics [1].

Plausible consequences include prolonged valuation pressures for private and newly public artificial intelligence enterprises, increased pricing power for enterprise software buyers, and a heightened emphasis on operational profitability over pure volume expansion across the technology sector.

Uncertainties remain regarding whether emerging use cases, such as personal agents, will generate sufficient economic productivity to recreate the hypergrowth surge previously driven by AI coding tools.