AI agent startup Instinct raises $1 billion at $10 billion valuation to build personal AI agent

AI agent startup Instinct has raised $1 billion at a $10 billion valuation, giving the San Francisco company a massive war chest to pursue one of AI’s biggest consumer bets: an assistant that does more than answer questions and actually gets things done.

The Series C round drew backing from Sequoia Capital, Benchmark Capital, and Coatue, Instinct announced Monday.

At the center of the bet is a personal AI agent built for the small, time-consuming jobs that fill everyday life. Instinct can plan trips, order groceries, cancel forgotten subscriptions, make reservations, place phone calls, and handle other tasks end to end.

Users communicate with Instinct by text or phone. The agent then uses its own computer and phone to carry out the requested job, pushing the familiar chatbot interface into something closer to a digital operator working on a user’s behalf.

That distinction has become one of the next major battlegrounds in AI. Chatbots taught millions of people to ask AI for information. Agent companies are betting users will increasingly ask AI to take action.

Instinct founder Noah Shinn is aiming squarely at that shift.

“We’re building Instinct to be the best personal agent that can handle the deeply personal nuances of everyday life,” Shinn said. “This funding helps us bring Instinct to more people and continue building the future of personal AI. It’s an exciting, creative time, and we’re just getting started.”

From chatbot to personal operator

Instinct remains in early access, yet the company has already started stretching the product beyond basic assistant tasks.

Its Instinct Concierge service tackles jobs that still require human-style persistence, such as calling a restaurant that does not accept online reservations, joining a dentist’s cancellation list, booking higher-end services, or dealing with a cable bill.

The company has introduced a Trusted Person Network, built around an Instinct-to-Instinct communication protocol. It lets one user’s assistant communicate with another user’s assistant to coordinate plans and exchange files such as PDFs, images, and spreadsheets.

Location sharing adds another layer. When shared through iMessage, Instinct can recognize when a user arrives at a location and respond with suggestions or information based on personal preferences.

That level of access raises an obvious question for any personal agent: how much authority should software have over someone’s accounts, credentials, communications, and private data?

Instinct says it built isolated sandboxes, short-lived local credentials, and identity-signed tool execution into the product from the beginning. The company has introduced what it calls an “active detective system” to identify hallucinations before the agent produces or executes a response.

The $1 billion round shows just how much capital investors are willing to put behind the idea that AI agents could become a new consumer computing layer.

The harder part comes next.

Building an assistant that can answer a question is one problem. Building one people trust to make calls, spend money, communicate with other agents, access personal information, and act in their name is a much higher bar.

Instinct now has $1 billion more to try to clear it.