Arcade.dev Raises $60M Series A as Enterprise AI Moves From Chat to Action

AI agents have reached an awkward stage of adolescence. They can write emails, summarize documents, analyze contracts, and generate software code. Yet the moment an organization asks an AI agent to actually do something inside a business system, the conversation changes. Security teams start asking questions. Compliance teams start making faces. Legal teams suddenly discover new calendar availability. That reality sits at the center of Arcade.dev's latest funding announcement.

Arcade.dev, a San Francisco-based company building what it calls the secure action layer for production AI agents, has raised $60M in Series A funding led by SYN Ventures, with participation from Morgan Stanley and Wipro. The round brings total funding to $72M, including the company's previous $12M seed round. According to the company, the new capital will support product development, ecosystem expansion, and hiring. The funding signals a broader shift occurring across enterprise AI, where the conversation is moving beyond whether AI can generate useful outputs and toward whether AI can securely take action inside real organizations without creating operational, security, and governance headaches.

What Happened

Arcade.dev announced a $60M Series A led by SYN Ventures, with strategic participation from Morgan Stanley and Wipro. The company was founded by Alex Salazar, Co-Founder & CEO, and Sam Partee, Co-Founder & CTO. Their focus is not building another foundation model or competing in the increasingly crowded race for AI intelligence. Instead, Arcade.dev is tackling a problem that emerges after the demo ends, when AI agents need permission to interact with applications, databases, APIs, internal systems, and workflows while operating within clearly defined authorization boundaries.

Enterprises need visibility into every decision, every request, and every system touched along the way. Arcade.dev positions itself as the infrastructure layer that makes those interactions secure, auditable, and manageable. The company operates within the emerging AI agent infrastructure category, which focuses on security, governance, authorization, and enterprise deployment. Arcade.dev reports 25x growth in tool-call volume over the past 6 months and says it has built more than 8,000 MCP tools designed for delegated user authorization. According to the company, the platform is already operating in production environments that include a top U.S. bank, Prosus, and LangChain.

As part of the funding round, Jay Leek, Managing Partner at SYN Ventures, joined Arcade.dev's board, adding another layer of enterprise cybersecurity expertise as the company scales its presence across large organizations.

Why This Matters

The AI industry spent the past few years obsessed with intelligence. The next phase will be defined by execution. A language model generating an answer creates relatively limited risk, but an AI agent with permission to update financial records, approve transactions, modify customer accounts, access internal systems, or trigger operational workflows introduces an entirely different set of challenges. This is where many enterprise AI initiatives encounter friction, because organizations quickly discover that model performance is only one piece of the equation.

Governance, authorization, compliance, security, and operational oversight become equally important once AI agents move into production environments. Arcade.dev is targeting that gap. The company's focus reflects a larger trend unfolding across enterprise software, where businesses increasingly care less about whether an AI model can generate text and more about whether an AI system can reliably perform tasks inside existing workflows. That distinction may end up defining the next generation of AI infrastructure winners.

Market Context

Enterprise AI adoption is entering a more mature phase. Early deployments focused heavily on experimentation through chatbots, copilots, and internal assistants. Many generated promising results, but fewer successfully scaled into mission-critical operations. The constraint was rarely intelligence. The challenge was trust. Enterprise environments contain sensitive customer information, regulated data, proprietary systems, financial records, and operational processes, making unrestricted AI access a nonstarter for security and compliance teams.

This reality has created growing demand for infrastructure that sits between AI agents and enterprise systems. Arcade.dev is operating within a category that includes AI governance, AI authorization, agent infrastructure, and enterprise security. These categories are becoming increasingly interconnected as organizations move from experimentation toward production deployment. Enterprise spending on AI infrastructure, governance, and security continues to rise as organizations push AI systems deeper into business operations.

Competitive Landscape

The AI market has no shortage of companies focused on making models smarter. Far fewer are focused on helping those models operate safely at scale. That creates an interesting dynamic for Arcade.dev. The company is not competing directly with OpenAI, Anthropic, Google, or other model providers. Instead, it sits downstream from those platforms, helping enterprises manage how AI agents interact with systems, applications, and users.

Arcade.dev also occupies an influential position within the emerging Model Context Protocol ecosystem. Model Context Protocol (MCP) is an emerging framework that allows AI agents to securely connect with tools, applications, and enterprise systems. Arcade.dev authored the MCP authorization specification adopted by Anthropic and continues contributing to standards development through the Linux Foundation's Agentic AI Foundation. Standards work rarely generates headlines, but it often shapes how quickly entire technology markets can scale.

What This Signals

The investor roster may be the strongest signal in the announcement. SYN Ventures built its reputation around cybersecurity investments. Morgan Stanley operates within one of the world's most heavily regulated financial environments. Wipro works alongside large enterprises navigating digital transformation initiatives across industries. These organizations spend their time evaluating risk, operational complexity, and long-term technology adoption patterns.

Their participation suggests a growing belief that enterprise AI adoption will increasingly depend on infrastructure layers that manage authorization, governance, and operational trust. In other words, AI's next challenge may not be intelligence. It may be accountability. That creates a different market opportunity from the one that dominated the first wave of generative AI investing.

The Bigger Industry Shift

Technology markets tend to move in predictable cycles. First comes capability. Then comes adoption. Eventually comes control. Generative AI has already proven capability. Enterprise adoption is accelerating. The market is now entering the control phase, where organizations must determine how AI systems operate safely within complex business environments.

Arcade.dev's funding announcement sits directly at that intersection. The company is betting that enterprises will increasingly require infrastructure that connects AI intelligence with governance, authorization, and operational oversight. Recent customer growth, production deployments, and investor interest suggest that bet is attracting attention. The broader takeaway extends well beyond a single funding round. Enterprise AI is evolving from a conversation about what machines can generate into a conversation about what systems can safely execute, and that shift could define the next decade of enterprise technology.