When Your Customer Becomes Your Competitor Overnight

The Fight That Ended in a Draw

Late Wednesday evening, Runlayer and Rippling simultaneously withdrew their lawsuits against one another. No settlement was negotiated. No legal fees were exchanged. The court documents confirm a clean break, with both sides walking away empty-handed.

Within hours, Rippling released its MCP gateway product - the very system that sparked the entire dispute. The launch was immediate and deliberate, a signal that whatever constraints the legal battle might have imposed were now irrelevant.

At DailyTechWire, we've tracked dozens of competitive skirmishes in the enterprise software space, but this one stands out for its speed and its implications. The conflict lasted barely three weeks in discovery before both parties concluded that continuing was pointless. What remains is a vivid illustration of how fragile competitive moats have become in the age of generative AI.

The Genesis of the Dispute

Runlayer emerged from stealth in November 2025 with $42 million in funding from Khosla Ventures and Felicis. The company is led by Andrew Berman, a serial founder whose previous ventures include Nanit, a baby-monitor manufacturer, and Vowel, an AI-powered video conferencing tool acquired by Zapier in 2024.

The startup's core product is an MCP gateway - a security layer that mediates requests from AI agents to enterprise software systems. When an AI agent needs to retrieve candidate resumes from a recruitment platform or pull financial records from an ERP system, the gateway handles authentication, access control, and logging. It prevents agents from having direct, unmonitored access to sensitive systems.

Rippling, a payroll and benefits management platform valued in the billions, spent more than a year testing Runlayer's gateway. The two engineering teams collaborated closely during this extended trial period. Then, according to Runlayer's legal filings, a Rippling employee sent Berman a text message. The employee disclosed that Rippling was building its own MCP gateway and intended to launch it as a standalone product. The employee described it as a clone of Runlayer's system.

Runlayer sued, alleging breach of contract tied to the testing agreements. Rippling responded with a countersuit claiming patent infringement - a move widely interpreted as an attempt to escalate legal costs and force Runlayer to back down.

What an MCP Gateway Actually Does

For those outside the enterprise AI security space, the technical stakes here are worth unpacking. An MCP gateway sits between AI agents and the software infrastructure they need to query. Imagine a hiring manager asks an AI assistant for the top five candidates for a role, including contact details and interview scores. That data lives in a recruitment system, possibly a CRM, and perhaps a scheduling tool.

Without a gateway, the agent would need direct API access to each system - a security nightmare. With a gateway, the request is routed through a centralized layer that enforces role-based permissions, logs every query, and can inject additional security policies. A manager might see full candidate profiles; an intern might see only anonymized summaries.

The gateway also provides observability: IT teams can audit which agents are making which requests, spot anomalies, and shut down rogue agents that might be operating without authorization. In an era when employees are spinning up AI tools without IT approval, this kind of centralized control is increasingly valuable.

Why Rippling Built Its Own

Rippling's decision to build rather than buy is not unusual in enterprise software, but the timeline is striking. The company, historically focused on payroll, benefits, and HR workflows, has rapidly expanded into adjacent markets. Its new MCP gateway includes model routing - sending requests to different large language models based on cost or performance - and token spend dashboards that track AI usage by employee.

This positions Rippling in direct competition not only with Runlayer but also with Docker, Amazon Bedrock, and even horizontal players like Stripe and Ramp, which have launched their own AI infrastructure products. The speed with which Rippling developed and shipped the gateway underscores how commoditized certain classes of software have become.

Generative AI tools have compressed development cycles. What might have taken a team six months to build in 2022 can now be prototyped in weeks. This has profound implications for startups that rely on technical differentiation as their primary moat.

The Broader Warning for Founders

The Runlayer-Rippling saga is a microcosm of a larger shift in enterprise software dynamics. Long technical evaluations - once a standard part of enterprise sales - are now riskier for startups. During a months-long pilot, a prospect's needs can change dramatically. Worse, the prospect can decide to build the solution in-house, leveraging the insights gained during the trial.

This is not a new phenomenon. Salesforce, Microsoft, and Oracle have long histories of absorbing features from smaller competitors or partners. But the acceleration of product cycles means the window for startups to establish defensibility is narrowing. A year-long technical engagement is no longer a signal of imminent revenue; it's a period during which the prospect is learning, evaluating, and possibly replicating.

Founders in the AI infrastructure space are now wrestling with how to structure pilot agreements. Some are shortening trial periods. Others are embedding contractual clauses that limit what prospects can build internally during and after evaluations. A few are shifting to open-source models, betting that community adoption and support contracts will prove more defensible than proprietary code.

What Runlayer Does Next

Runlayer's positioning has always been broader than a single gateway product. The company offers a bundle of agent security services, including tools for creating agents, monitoring for shadow AI deployments, and enforcing enterprise-wide policies. The gateway is one component of a larger platform.

The legal fight, while brief, served as an unintended marketing event. Runlayer's name recognition among enterprise IT buyers likely increased. Whether that translates into customer traction remains to be seen. The company will need to demonstrate that its broader security bundle offers enough value to justify the cost, especially now that competitors like Rippling are offering gateways as part of existing subscriptions.

Rippling's Expanding Ambitions

Rippling's entry into AI security is part of a broader strategy to own more of the enterprise stack. The company already manages payroll, benefits, device management, and app provisioning. Adding an AI gateway allows it to extend that control into the agent layer, tying AI access to employee identity and role.

This vertical integration is appealing to enterprise buyers who prefer consolidated vendors. It also creates lock-in: once a company's HR, payroll, and AI infrastructure are all managed by Rippling, switching costs rise significantly. The company is betting that bundling will outweigh best-of-breed specialization, a bet that has paid off in other software categories.

The New Calculus for Enterprise AI Startups

The incident raises uncomfortable questions for early-stage companies targeting large enterprises. How do you de-risk long sales cycles when the buyer might become a competitor? How do you protect intellectual property during pilots without alienating prospects? And how do you build defensibility when code can be replicated in weeks?

Some investors are advising portfolio companies to focus on vertical niches where larger platforms are less likely to follow. Others are pushing startups to move upmarket faster, locking in contracts before competitors can mobilize. A third camp argues for open-source strategies, where the code is freely available but the startup retains expertise and integration advantages.

None of these strategies is foolproof. The reality is that in a world where software development is increasingly assisted by generative AI, technical moats are eroding. Startups must find defensibility elsewhere: in customer relationships, in data network effects, in regulatory expertise, or in operational excellence.

A Truce, Not a Resolution

The lawsuits are over, but the competitive tension is not. Rippling's MCP gateway is now live. Runlayer continues to sell its broader security platform. Both companies will compete for the same enterprise customers, many of whom are still figuring out how to govern AI agents in production environments.

The legal standoff cost both sides time, attention, and reputational capital. Neither emerged with a clear victory. What they did achieve, inadvertently, was a public lesson for other founders navigating the same terrain. In the age of AI, the line between customer and competitor is thinner than ever. And the speed at which that line can be crossed is only accelerating.