Robinhood launches beta support for AI agentic trading and payments

Robinhood is rolling out a beta feature called Agentic Trading that lets users hand the wheel to third-party AI agents for executing trades. The twist: it all happens inside a separate, pre-funded brokerage account, keeping your main portfolio safely out of the algorithm’s reach.

Users create a dedicated self-directed account, load it with funds, and then connect an external AI agent through what’s known as the Model Context Protocol, or MCP. The agent can then buy and sell equities, options, and crypto on the user’s behalf.

How the sandboxed trading works

MCP is a protocol designed to facilitate secure, isolated interactions between AI models and external systems, creating a walled garden where the AI agent can operate without ever touching the user’s primary account.

This builds on a capability Robinhood already had in place, namely the ability for users to open multiple individual brokerage accounts. That feature was originally designed to let people compartmentalize strategies. Agentic Trading takes the same infrastructure and repurposes it for AI-driven execution.

The feature supports trading across equities, options, and cryptocurrencies.

Robinhood’s broader AI push

Agentic Trading is part of a larger AI push that Robinhood has been building out since early 2025.

The company also launched Robinhood Cortex, an AI assistant that lets users manage investment strategies through natural language commands. The assistant handles both equity and cryptocurrency transactions. Cortex was launched in March 2025.

Cortex is Robinhood’s own AI, built into the platform and designed to follow user instructions in real time. Agentic Trading, by contrast, opens the door to external AI agents, meaning third-party models that users choose and configure themselves.

Robinhood has published comprehensive support documentation for the beta. The beta rollout began in March 2025.

What this means for investors

Connecting third-party AI agents to live brokerage accounts introduces a layer of complexity that most retail investors have never dealt with. The sandboxed account structure mitigates the worst-case scenario of an agent draining a primary portfolio, but it does not eliminate the risk of losses within the dedicated account itself.

Disclosure: This article was edited by Editorial Team. For more information on how we create and review content, see our Editorial Policy.