OpenAI Invests in Isara, a Startup Building AI Agent Swarms at $650 Million Valuation

Funding and Founders

Isara, a San Francisco startup founded nine months ago by Eddie Zhang and Henry Gasztowtt, announced a $94 million financing round that values the company at $650 million. The round was first reported by The Wall Street Journal and includes OpenAI, Amity Ventures, Michael Ovitz (former Creative Artists Agency chairman and early Uber backer) and billionaire hedge‑fund manager Stanley Druckenmiller. Zhang, a former OpenAI AI safety researcher, left the company in June 2025 to launch Isara with Gasztowtt, a computer‑science student at the University of Oxford. The duo previously co‑authored a paper at ICML 2024 on AI cooperation for policymaking, which serves as the intellectual foundation for the startup.

Technology Vision

Isara’s core ambition is large‑scale multi‑agent coordination. While most current AI products rely on a single model responding to a single prompt, Isara’s architecture is designed to enable hundreds or thousands of specialized agents to communicate, divide tasks, align on goals and produce a combined output. The company describes this shift as moving from isolated tools to coordinated teams of agents.

In its initial demonstration, roughly 2,000 agents worked together to forecast the price of gold. The startup’s first commercial target is investment firms that would pay for predictive‑modeling software. Secondary markets include biotechnology and geopolitical analysis, and the long‑term vision is to train agent swarms to track geopolitical shifts and forecast economic trends.

Technical Challenges

Coordinating thousands of agents presents significant hurdles. Even a single AI agent can struggle with complex tasks; scaling to thousands raises concerns about cascading errors, conflicting objectives and compounded hallucinations. Existing multi‑agent frameworks such as LangChain, CrewAI and AutoGen typically coordinate small numbers of agents on relatively structured tasks, making Isara’s goal a different order of difficulty.

Industry Context

Isara is part of a broader pattern labeled “neolabs” by The Information—research‑heavy AI startups founded by alumni of OpenAI, DeepMind, Anthropic and Google Brain that operate more like privately funded research institutions than traditional companies. Investors have poured billions into such startups, betting that breakthroughs will come from architectures fundamentally different from the large language models that dominate the market today.

The investment thesis focuses on foundational research capability rather than immediate revenue. While companies like Cognition have demonstrated strong revenue growth alongside high valuations, the upside from a genuine breakthrough in multi‑agent coordination could be enormous, albeit uncertain.

OpenAI’s Strategic Rationale

OpenAI’s participation raises strategic questions. By investing $94 million at a $650 million valuation, OpenAI gains optionality in a technology area adjacent to its own research agenda. The investment also helps retain talent; maintaining a relationship with Zhang and his team provides OpenAI with insight into emerging approaches without losing the researchers to competitors.

Outlook

Isara now has roughly a dozen additional researchers hired from Google, Meta and OpenAI. The next 18 months will test whether the company can translate its research vision into reliable, revenue‑generating products. The agentic AI market is projected to grow from $7.8 billion in 2025 to $52.6 billion by 2030, and major platforms are already shipping multi‑agent features. Isara’s bet is that coordination at the scale of thousands of agents requires a fundamentally different architecture than what incumbents currently offer.