The AI Arms Race: How Ramp and FIS Are Shaping the Future of Corporate Finance and Security

By Tatiana Walk-Morris
Published July 22, 2026

As the global financial sector grapples with the transformative, yet unpredictable, nature of generative artificial intelligence, two prominent players in the payments landscape—Ramp and FIS—have unveiled distinct strategies to manage the risks associated with this technology. While one firm focuses on the escalating financial burden of AI consumption, the other is doubling down on the security infrastructure required to protect the integrity of digital payments.

These moves, announced this July, underscore a broader trend: artificial intelligence is no longer just a "value-add" for fintech companies; it is a critical operational variable that requires sophisticated governance, oversight, and defense mechanisms.

Main Facts: Managing Costs and Fortifying Perimeters

The rapid adoption of AI has created a dual-front challenge for enterprise leaders: "sticker shock" from ballooning subscription costs and an expanded attack surface for cyber-criminals.

Ramp, the New York-based spend management platform, has identified a critical pain point for modern CFOs: the lack of visibility into "AI token consumption." As employees across departments sign up for varied large language model (LLM) services—ranging from OpenAI and Anthropic to Google’s Gemini—spending has become decentralized and difficult to track. Ramp’s new tool is designed to provide centralized governance, allowing finance teams to monitor, categorize, and cap AI expenditures in real-time.

Simultaneously, Jacksonville-based financial services technology giant FIS is addressing the systemic threat that AI poses to the global payments ecosystem. By joining Anthropic’s "Project Glasswing"—a collaborative cybersecurity initiative—FIS is positioning itself at the forefront of AI-driven threat detection. This move aligns with FIS’s ongoing commercial partnership with Anthropic, integrating advanced AI agents to bolster security protocols across its vast network.

Chronology of the AI Integration Pivot

The trajectory of these initiatives reflects a fast-moving industry environment where reactive measures are quickly being replaced by proactive, integrated governance.

• Early 2024 – Mid 2025: Companies across the financial services sector began experimenting with LLMs to automate customer support, coding, and document analysis. During this period, procurement departments were often bypassed as individual teams experimented with AI tools on corporate credit cards.
• March 2026: At major industry trade fairs in Europe and North America, a consensus emerged: AI, while efficient, was becoming a "black box" for enterprise finance and a "wild west" for security teams.
• July 2026: Ramp formally launches its AI spend-tracking functionality, responding to a surge in customer complaints regarding unpredictable billing cycles from AI providers.
• July 2026: FIS announces its formal integration into Project Glasswing, marking a significant escalation in its commitment to collaborative, AI-led cybersecurity.

Supporting Data: The Cost of the AI Explosion

The data underpinning Ramp’s new initiative is stark. According to the company’s internal metrics, spending on AI tokens has become the fastest-growing category of business expenditure. Unlike traditional SaaS subscriptions, which often follow a predictable per-user pricing model, AI usage is frequently tied to consumption metrics—or "tokens."

For a large enterprise with thousands of employees, this means a single project’s budget can be obliterated in days if an automated process goes haywire or if a team inadvertently exceeds their usage threshold. Ramp’s data reveals that finance teams currently have "little to no visibility" into this spend, leading to significant budget variances and, in some cases, the suspension of AI projects due to unforeseen costs.

On the security front, the urgency is equally data-driven. With the rise of sophisticated AI-generated phishing and synthetic identity fraud, traditional rule-based security measures are becoming obsolete. The collaboration between firms like FIS and developers like Anthropic represents a move toward "AI-vs-AI" defense, where machine learning models are tasked with identifying patterns of fraud that occur at speeds invisible to human analysts.

Official Responses and Strategic Rationale

Ramp: Bringing Governance to Chaos

Ramp’s strategy is rooted in the belief that "visibility equals control." In their official announcement, the company highlighted that the new tool allows for granular control: users can set spending limits by department, project, or individual employee. Furthermore, the system provides automated alerts for sudden surges in usage, preventing the "cost overruns" that have plagued firms over the last 18 months.

"Spending on AI tokens has become the fastest growing category in business spend, but finance teams have little to no visibility into how it’s being spent," a spokesperson for Ramp stated. By aggregating data across providers like OpenAI, Anthropic, and Gemini, Ramp is effectively creating a "dashboard for the AI era," turning opaque token usage into actionable financial intelligence.

FIS: A Proactive Security Stance

For FIS, the decision to participate in Project Glasswing is framed as a duty to the broader financial services sector. Project Glasswing, which counts tech giants like Google, Amazon Web Services, and JPMorgan Chase among its participants, is an effort to share threat intelligence and develop defensive AI models that can anticipate and neutralize malicious activity.

"This is a commitment to proactive security and being a supportive partner to the broader security community and financial services sector," FIS noted in its statement. By leveraging Anthropic’s models, FIS is not only protecting its own infrastructure but is also contributing to a collective defense strategy that is essential in an era where cyber threats are becoming increasingly automated.

Implications: The New Standard for Financial Services

The dual moves by Ramp and FIS signal a maturation of the AI market. We are moving past the "enthusiasm phase" and into the "industrialization phase."

1. The Rise of "AI Spend Management" as a Category

Ramp’s initiative may well set a new industry standard. As AI becomes embedded in workflows, CFOs will likely demand that AI token consumption be treated with the same level of scrutiny as travel and entertainment (T&E) or cloud computing infrastructure costs. We should expect other fintechs to follow suit, launching their own "AI governance" modules in the coming quarters.

2. Collaborative Defense

FIS’s participation in Project Glasswing illustrates the limitations of "siloed security." Because AI-driven threats often scale horizontally across the industry, no single institution can protect itself in isolation. The shift toward industry-wide, AI-powered security cooperatives suggests that the future of banking security will be defined by data-sharing agreements and cross-firm collaboration.

3. The Human Element

Despite the heavy focus on automation, both initiatives highlight the persistent need for human oversight. Ramp’s tool requires managers to set limits and review alerts; FIS’s security integration relies on the guidance of cybersecurity professionals to interpret the outputs of its defensive AI agents. The role of the human in the loop is not being replaced; it is being elevated to a higher level of architectural oversight.

4. Regulatory Pressures

As AI spending and security vulnerabilities enter the boardroom agenda, regulators are likely to take notice. Transparency in AI spending and the robustness of AI-based security protocols could soon become standard audit requirements for publicly traded financial services companies. Firms that adopt robust governance models early, as Ramp and FIS have done, will be better positioned to navigate the regulatory scrutiny that is undoubtedly on the horizon.

Conclusion: A Balanced Approach

The developments at Ramp and FIS represent the two sides of the same coin. One is focused on the economics of AI—ensuring that the business case for AI remains viable by preventing runaway costs—while the other is focused on the resilience of AI—ensuring that the technology does not become a conduit for catastrophic risk.

As we look toward the remainder of 2026, it is clear that the companies that thrive will be those that can successfully manage the inherent volatility of the AI revolution. Whether through the granular tracking of token consumption or the collaborative defense of payment networks, the message to the market is consistent: AI is a powerful tool, but it is one that requires constant, vigilant management. The "AI changes everything" sign seen at the Berlin trade fair is no longer a marketing slogan—it is a reality that is fundamentally rewriting the playbook for modern finance.