AI security startup Upwind raises $400 million at a $3.8 billion valuation

Investors are putting large sums into companies that protect against a new kind of risk: AI systems that act on their own.

What happened

Upwind, a cloud and AI security company, has raised $400 million at a valuation of $3.8 billion. The company says its revenue quadrupled in 2026. It is focusing on risks from autonomous AI agents.

Background

AI agents are programs that do more than answer questions. They can take actions: running code, calling online services, moving files or making purchases, often with access to company systems and data.

That power creates new security problems. An agent may be tricked by hidden instructions in a web page or document, a technique known as prompt injection. It may be given more access than it needs, or its actions may be hard to track. Traditional security tools were built to watch people and ordinary software, not programs that make their own decisions.

Cloud security firms like Upwind monitor the systems where companies run their applications and data, looking for misconfigurations, vulnerabilities and suspicious activity.

Why it matters

The size of the round shows how seriously businesses are taking AI risk. As companies hand more tasks to agents, a single compromised agent could expose data or cause damage at machine speed.

The same concerns apply to crypto. Agents that can sign transactions or manage wallets would be prime targets, and the security tools being built now will shape how safely that can happen.