HR Giant Rippling Accused of Cloning AI Startup's Tech After Year-Long Trial
Key takeaways
- Runlayer alleges Rippling used a year-long product trial to copy its MCP gateway source code and IP, despite a signed non-disclosure and trial agreement
- A 'Rippling insider' allegedly tipped off Runlayer's CEO that an internal project was building a near-identical clone of the startup's product
- Rippling denies all wrongdoing and confirms it is launching its own MCP gateway, calling Runlayer's lawsuit an attempt to avoid legitimate competition
A legal clash between two tech companies is shining an uncomfortable spotlight on one of enterprise software's most dangerous vulnerabilities: the product trial. Runlayer, a startup building secure Model Context Protocol gateways — technology that lets AI agents safely connect to external data and tools — has filed a lawsuit against HR software giant Rippling, alleging that a months-long evaluation process was used as cover to replicate its product from the ground up. The complaint was reviewed by TechCrunch and lays out a detailed and alarming sequence of events.
According to the lawsuit, Rippling approached Runlayer as a prospective customer and the two companies entered into a mutual non-disclosure agreement, along with a product trial agreement that explicitly prohibited Rippling from copying Runlayer's intellectual property or creating derivative works. Despite these protections, Runlayer alleges that Rippling's evaluation stretched into what the complaint calls 'nearly a year of intensive engineering collaboration,' during which Runlayer shared not just its product but also its roadmap and actual source code. When the two sides ultimately couldn't agree on pricing, Runlayer terminated the trial.
Shortly after the trial ended, Runlayer founder and CEO Andrew Berman allegedly received a text from a 'Rippling insider' warning him that the company had launched an internal project to build what was described as 'almost a 1 to 1 copy of Runlayer.' The startup's legal team — the white-shoe firm Sullivan & Cromwell, which itself signals how seriously Runlayer is taking this fight — argues the copycat product constitutes trade secret misappropriation, unfair competition, and breach of contract. Rippling has confirmed it is launching its own MCP gateway but flatly denies misusing Runlayer's IP, calling the lawsuit a 'panicked effort to avoid competition.'
Runlayer is not operating in a niche corner of the market. Anthropic open-sourced the MCP protocol in November 2024, and it has since become a foundational building block for AI interoperability. MCP gateways layer on additional enterprise features like access control, security monitoring, and agent management. Runlayer launched its own gateway product in mid-2024 and has raised $42 million in total funding from investors including Khosla Ventures and Felicis. But the space has grown considerably more competitive since then, with well-resourced companies increasingly capable of building similar tools in-house.
The lawsuit captures a tension that many AI infrastructure startups face but rarely discuss openly. Enterprise sales cycles are notoriously long and require deep technical transparency to close, yet that very transparency creates exposure. When a deal falls through — especially over something as routine as a pricing disagreement — the startup is left having handed over its most sensitive assets with nothing to show for it. The Runlayer case may not be the last of its kind, and its outcome could meaningfully influence how AI startups structure their product trials going forward.
The bigger picture
The Runlayer versus Rippling lawsuit exposes something the AI infrastructure sector has been quietly grappling with: the closer you get to selling to a well-funded tech company, the more you risk becoming their R&D department for free. Enterprise trials, by design, require deep technical integration and knowledge transfer. For a small startup, that's a calculated bet. For a large company with hundreds of engineers on staff, the calculus can shift quickly once they understand the architecture well enough to replicate it. This dynamic has always existed in enterprise software, but the speed and modularity of modern AI tooling makes it acutely dangerous.
Rippling's framing — that Runlayer is using litigation to dodge competition — is also worth scrutinizing. It's a standard counter-narrative, and it may even be partially true that Rippling built its product independently. But the specific allegations here, including the whistleblower text and the timing of Rippling's internal project relative to the trial's end, are harder to dismiss as coincidence. Courts will have to weigh evidence, not optics. The presence of Sullivan & Cromwell suggests Runlayer has at least some documentary evidence it believes is compelling, and Rippling's aggressive public response suggests it takes the reputational risk seriously.
For the broader AI startup ecosystem, this case is a warning shot. Founders building infrastructure tools should revisit how they structure evaluations — what they share, when they share it, and what technical watermarking or access logging they implement to create an evidentiary trail. The MCP gateway space in particular is becoming commoditized faster than anyone anticipated, and startups that survive will need both differentiated product and airtight legal architecture. Watch this case closely: a favorable ruling for Runlayer could reshape enterprise trial norms across the industry.
We decided to cover this story at LagPing because it sits at the intersection of two things we track closely: the fast-moving AI infrastructure space and the business risks that founders rarely talk about publicly until something goes wrong. Runlayer's situation is not an isolated incident — it's a symptom of a structural problem in how AI startups sell to enterprise customers, particularly those with strong internal engineering teams. The MCP protocol itself is something we've been watching since Anthropic released it in late 2024, and the emerging market of gateway products built on top of it is exactly the kind of evolving landscape our readers care about. Beyond the legal drama, this case raises genuinely important questions about how the next generation of AI infrastructure companies should protect themselves during commercial negotiations. We think it deserves more than a headline — it deserves context, analysis, and a frank conversation about what it means for founders building in this space right now.
As an Amazon Associate, LagPing earns from qualifying purchases. Product links are affiliate links.
You might also like
Rippling's CFO Shock: How Runaway Token Bills Spawned a New AI Cost-Control Product
15h ago
Music Giants Want AI Tracks Barred From Charts Unless They Meet Humanity Test
Aug 1
Tech Giants Form AI Defense Pact After Rogue Model Breach Exposed Critical Security Gaps
Jul 28
Claude Opus 5 Arrives Boasting Near-Fable 5 Power Amid a Turbulent Season for AI Giants
Jul 25