AI Infrastructure Battle Heats Up: Runlayer Sues Rippling Over Alleged IP Theft After Failed Product Trial

A high-stakes legal battle is unfolding in the burgeoning artificial intelligence infrastructure sector, as Runlayer, a startup specializing in secure Model Context Protocol (MCP) gateways, has filed a lawsuit against HR software giant Rippling. The complaint, obtained by TechCrunch, accuses Rippling of trade secret misappropriation, unfair competition, and breach of contract, alleging that Rippling cloned Runlayer’s technology after an extensive product trial. This litigation serves as a stark warning for technology companies, particularly those selling complex AI infrastructure to large enterprises that often possess the internal engineering capabilities to replicate solutions in-house.

The Core Allegations: A Year-Long Collaboration Ends in Accusation

At the heart of Runlayer’s lawsuit is the claim of a protracted and intimate product trial with Rippling, spanning nearly a year of intensive engineering collaboration. During this period, Runlayer, whose core offering is a gateway that enables AI models and agents to securely access external data and tools, asserts it shared a trove of proprietary information with Rippling. This included sensitive details such as its product roadmap and, critically, its actual source code. To facilitate this deep engagement, both parties reportedly entered into a mutual non-disclosure agreement (NDA), a standard practice in enterprise-level technology evaluations. Furthermore, Rippling signed a specific product trial agreement that explicitly prohibited it from copying Runlayer’s intellectual property (IP) or creating derivative works—a common boilerplate clause designed to protect vendors during extensive evaluations.

Runlayer, founded by CEO Andrew Berman, maintains that this period of collaboration was not merely a superficial demonstration but a profound technical partnership. Its engineers worked closely with Rippling’s teams, providing insights and access that are typically reserved for established partnerships or direct product integration. The objective, from Runlayer’s perspective, was to demonstrate the full capabilities and robustness of its MCP gateway, paving the way for a significant enterprise licensing agreement. However, despite the depth of the technical engagement, the two companies ultimately failed to reach an agreement on pricing, leading Runlayer to terminate the product trial.

The situation escalated dramatically shortly after the trial concluded. According to the complaint, Berman received an alarming text message from an alleged "Rippling insider." This individual reportedly informed Berman of an internal project at Rippling aimed at building "essentially a clone o[f] Runlayer… it’s almost a 1 to 1 copy of Runlayer." This alleged communication forms a cornerstone of Runlayer’s argument, suggesting direct evidence of Rippling’s intent and execution in replicating its technology. Runlayer contends that Rippling’s subsequent launch of its own MCP gateway product must, by necessity, be based on the intellectual property it had disclosed during the trial, thus constituting a clear violation of their agreements and established legal protections for trade secrets.

Rippling’s Firm Denial and Counter-Narrative

In response to the lawsuit, Rippling has acknowledged its plans to launch its own Model Context Protocol (MCP) gateway. However, a spokesperson for Rippling vehemently denied Runlayer’s allegations of misusing its intellectual property. The company has framed the lawsuit as a desperate attempt by Runlayer to stifle legitimate competition, asserting that Runlayer’s claims are fabricated in the wake of its own "business failures."

“Runlayer’s panicked effort to avoid competition by fabricating claims is not an effective way to deal with its business failures,” a Rippling spokesperson told TechCrunch. “Rippling is launching a superior product for connecting AI tools to business data using only our proprietary information—we have every reason to win in this market.” This statement underscores Rippling’s position: that its internal development of an MCP gateway is independent, based solely on its own innovative efforts and proprietary data, and not on any misappropriated information from Runlayer. Rippling’s counter-narrative suggests that Runlayer, facing a competitive market and potentially unable to secure a lucrative deal, is now resorting to legal action to impede a rival’s legitimate market entry.

A Chronology of Engagement and Escalation

To fully understand the context of this dispute, it is crucial to establish a timeline of events and the broader market landscape.

Mid-2023: Runlayer officially launches its secure Model Context Protocol (MCP) gateway product. The company quickly gains traction, securing substantial funding, including an initial $11 million from investors like Khosla Ventures and Felicis Ventures, eventually raising a total of $42 million. This early investment signals strong confidence in Runlayer’s technology and its potential in the evolving AI infrastructure market.
Late 2023 – Late 2024 (Approx.): Runlayer and Rippling initiate discussions for a potential partnership or licensing agreement. This leads to the commencement of an extensive product trial. During this period, Runlayer shares its product roadmap and source code under mutual NDAs and a specific product trial agreement with IP protection clauses. The trial reportedly involves "nearly a year of intensive engineering collaboration," indicating a deep integration and technical exchange between the two companies’ development teams.
Late 2024: Negotiations between Runlayer and Rippling regarding pricing fail to reach a consensus. Runlayer subsequently terminates the product trial, ending the formal engagement.
Shortly After Trial Termination (Late 2024 / Early 2025): Runlayer CEO Andrew Berman allegedly receives a text message from a "Rippling insider," detailing an internal project to build a near-identical clone of Runlayer’s product.
November 2024: Separately, Anthropic, a prominent AI research company, launches its own Model Context Protocol (MCP) as an open-source standard. This development further intensifies the competitive landscape for MCP solutions, making the market more accessible and potentially challenging for proprietary offerings.
Early 2025 (Implied): Rippling confirms its intention to launch its own MCP gateway, which Runlayer alleges is the "clone" developed using its misappropriated IP.
Recent Past: Runlayer files a lawsuit against Rippling, alleging trade secret misappropriation, unfair competition, and breach of contract, seeking injunctive relief and damages. Runlayer retains the services of the prestigious "white-shoe" law firm Sullivan & Cromwell, lending significant credibility to its legal pursuit.

The Strategic Importance of Model Context Protocols (MCP) Gateways

The technology at the center of this dispute, the Model Context Protocol (MCP) gateway, is not merely a niche component but a foundational element in the secure and effective deployment of enterprise AI. As AI models and agents become increasingly sophisticated and integrated into business operations, their ability to securely access and interpret external, real-time data is paramount. Traditional methods of data integration often fall short in meeting the stringent security, privacy, and control requirements of large organizations.

An MCP gateway acts as a secure intermediary, providing a standardized, controlled, and auditable pathway for AI models to pull in information from various external data sources and interact with other tools. This is crucial for several reasons:

Security and Compliance: Enterprises deal with vast amounts of sensitive data. An MCP gateway ensures that AI interactions with this data adhere to strict security protocols, access controls, and regulatory compliance standards (e.g., GDPR, HIPAA). It minimizes the risk of data leakage or unauthorized access by AI agents.
Control and Governance: In complex enterprise environments, managing which AI models can access what data, and under what conditions, is vital. MCP gateways provide a centralized point of control for IT and security teams, allowing them to define policies, monitor usage, and revoke access as needed.
Interoperability: AI systems often need to interact with a diverse ecosystem of internal and external applications, databases, and APIs. An MCP standard facilitates this interoperability, enabling seamless data flow and tool utilization without requiring bespoke integrations for every new AI application.
Agent Management: As autonomous AI agents become more prevalent, managing their interactions with external systems becomes a significant challenge. MCP gateways offer features specifically designed for agent orchestration, ensuring they operate within defined parameters and do not pose unforeseen risks.

Runlayer’s value proposition revolved around providing a robust, secure, and enterprise-grade MCP gateway solution, designed to accelerate the safe adoption of AI within large organizations. Its significant funding rounds underscore the perceived market need for such a specialized solution.

Rippling’s AI Ambitions: Why an HR Tech Company Needs an MCP Gateway

Rippling, primarily known for its comprehensive HR, IT, and finance software platform, might seem an unlikely candidate to develop a core AI infrastructure component like an MCP gateway. However, its strategic move into this area reflects a broader trend in enterprise software: the deep integration of AI capabilities across all business functions.

Rippling’s platform automates various aspects of employee management, from payroll and benefits to IT provisioning and expense management. Integrating AI into such a platform offers immense potential for efficiency, personalization, and predictive analytics. For instance:

Automated HR Tasks: AI can streamline onboarding, answer employee queries, and even predict potential employee churn.
Personalized Employee Experiences: AI can tailor benefits recommendations, learning paths, or career development opportunities.
Intelligent IT Management: AI can proactively identify IT issues, automate software provisioning, and enhance cybersecurity.

To achieve these advanced AI functionalities, Rippling’s internal AI models would need secure and controlled access to a vast array of data—employee records, financial data, performance metrics, IT usage logs, and potentially external market data for benchmarking or insights. An MCP gateway would be critical for Rippling to manage this data flow securely, maintain compliance, and ensure the responsible use of AI within its extensive platform. Building such a component in-house would grant Rippling complete control over its AI data security architecture, potentially offering a competitive edge and tighter integration with its existing services.

The "Build vs. Buy" Conundrum: A Cautionary Tale for All

This lawsuit brings into sharp focus one of the most enduring and complex strategic dilemmas in the tech industry: whether to "build" a technology in-house or "buy" it from an external vendor. For AI infrastructure, this decision carries even greater weight due to the rapid pace of innovation, the specialized expertise required, and the high stakes involved in data security and performance.

For Enterprises (like Rippling):

Benefits of Building: Greater control, full customization, potential for deeper integration with existing systems, long-term cost savings (by avoiding recurring licensing fees), and the ability to develop a proprietary competitive advantage.
Risks of Building: Significant upfront investment in R&D, potential delays, difficulty attracting specialized talent, and the risk of diverting resources from core business activities.
The Trial Trap: Extensive product trials allow enterprises to thoroughly evaluate a solution, often gaining deep insights into its architecture and functionality. This knowledge, while essential for informed decision-making, can inadvertently facilitate an in-house build, especially if the enterprise has strong engineering capabilities and decides against the vendor’s pricing.

For Vendors (like Runlayer):

Benefits of Selling: Access to large enterprise customers, significant revenue potential, market validation, and the ability to scale specialized solutions across multiple clients.
Risks of Selling (during trials): The inherent risk of exposing sensitive intellectual property during deep evaluations. Vendors are caught between the need to demonstrate their product’s full capabilities (often requiring sharing core IP) and the fear of being copied. If a trial ends without a sale, the vendor may have inadvertently provided a blueprint for a competitor.
The IP Protection Challenge: Even with robust NDAs and trial agreements, proving IP theft can be incredibly challenging, especially when the alleged copying involves complex software and subtle architectural similarities rather than direct code plagiarism.

The Runlayer-Rippling case vividly illustrates this tightrope walk. Runlayer, eager to secure a major enterprise client, engaged in a deep collaboration, sharing its most valuable assets. Rippling, a tech company with considerable engineering muscle and a clear strategic interest in AI, gained unparalleled access. When pricing became a sticking point, the relationship dissolved, and now allegations of IP theft have emerged.

The Crowded and Evolving Landscape of AI Interoperability

The market for MCP gateways and similar AI interoperability solutions is becoming increasingly competitive and dynamic. Runlayer was an early mover, launching its product in mid-2023 and securing significant funding, reflecting its leadership in addressing this critical need. However, the landscape has evolved rapidly.

Open-Source Alternatives: Anthropic’s decision to launch MCP as an open-source protocol in November 2024 is a game-changer. Open-source initiatives can democratize access to foundational technologies, potentially lowering barriers to entry for companies wanting to build their own AI infrastructure. While proprietary solutions like Runlayer’s offer enhanced features, security, and enterprise support, the availability of a robust open-source alternative can pressure pricing and market share for commercial offerings.
Cloud Provider Offerings: Major cloud providers (AWS, Azure, Google Cloud) are rapidly expanding their AI infrastructure services, often including tools for secure data access and model deployment that could overlap with MCP gateway functionalities.
Other Startups and Integrators: A growing number of startups are entering the AI infrastructure space, offering specialized solutions for data integration, security, and AI lifecycle management.

This intensifying competition means that companies like Runlayer must not only innovate but also navigate the complex commercial landscape, where potential customers can quickly become competitors. Rippling, by potentially leveraging its internal capabilities and the general availability of open-source frameworks, could argue it is simply participating in a competitive market, distinct from any alleged IP theft.

Legal Ramifications and Broader Industry Implications

Runlayer’s decision to retain Sullivan & Cromwell, a "white-shoe" law firm renowned for its litigation prowess, signals its serious intent and financial capacity to pursue this case aggressively. While retaining a marquee law firm doesn’t guarantee victory, it certainly lends credibility and substantial legal firepower to Runlayer’s claims, both optically and practically.

The lawsuit will likely hinge on several complex legal determinations:

Definition of Trade Secrets: What specific information shared by Runlayer qualifies as a legally protected trade secret? Was it sufficiently novel, non-obvious, and kept confidential?
Misappropriation: Can Runlayer prove that Rippling actually used its trade secrets to build its own product? This often involves forensic analysis of code, architectural similarities, and internal communications. The alleged "insider" text could be a critical piece of evidence here.
Breach of Contract: Were the NDA and product trial agreement clear and enforceable? Did Rippling’s actions constitute a direct violation of the IP protection clauses?
Unfair Competition: Did Rippling’s alleged actions give it an unfair advantage in the market, beyond legitimate competitive practices?

If Runlayer prevails, the potential remedies could include monetary damages (e.g., lost profits, unjust enrichment), injunctive relief (forcing Rippling to cease development or use of the alleged clone), and even punitive damages.

Beyond the immediate parties, this case carries significant implications for the broader tech industry, particularly for:

AI Startups: It reinforces the paramount importance of robust legal agreements and careful IP protection strategies, especially when engaging with larger, well-resourced potential customers. It may lead to more restrictive trial agreements or staged disclosures of IP.
Enterprise Customers: It highlights the legal risks associated with "build vs. buy" decisions, especially after deep product evaluations. Companies must ensure their internal development efforts are demonstrably independent and do not infringe on the IP of former vendors.
The Future of AI Infrastructure: The outcome could influence how IP is protected and valued in the rapidly evolving AI sector, potentially shaping future engagement models between innovators and integrators. It underscores the value of proprietary solutions in a market increasingly influenced by open-source alternatives.

Conclusion: A Cautionary Tale Unfolding

The lawsuit between Runlayer and Rippling is more than just a dispute between two companies; it is a significant test case for intellectual property protection in the high-stakes world of enterprise AI. It exposes the delicate balance innovators must strike between demonstrating their cutting-edge technology to prospective clients and safeguarding their core assets from potential misappropriation. For enterprise customers, it serves as a powerful reminder of the legal and ethical obligations that accompany extensive product trials.

As the legal proceedings unfold, the tech industry will be watching closely. The outcome will not only determine the fates of Runlayer and Rippling but may also establish crucial precedents for how innovation is protected and commercialized in the next generation of artificial intelligence. It is a cautionary tale, still very much in the making, about the perils and promises of collaboration in a fiercely competitive technological landscape.