AI Can Move From Analysis to Order Submission
Interactive Brokers’ existing AI integration lets clients connect accounts to third-party AI platforms, generate trade instructions and review them before submission. Autonomous Trading adds the ability for an eligible connected system to place supported orders when predefined conditions are met. The distinction is operational: the AI is no longer limited to summarizing a portfolio or drafting an instruction for a person to approve.
Clients can provide instructions in natural language. Examples include monitoring implied volatility, evaluating a position after a decline from its recent high, trimming an oversized holding or submitting a covered-call order when a volatility condition is met. These examples describe user-defined triggers rather than recommendations from the broker. The system’s authority depends on what the client instructs it to do and what the account’s restrictions allow.
Chief Executive Milan Galik said the feature builds on the broker’s use of automation by allowing clients to define strategies and safeguards while connected AI monitors markets and executes within those parameters. The launch signals a change in how a brokerage can expose its order-routing and account functions to external AI systems. It also shifts some practical oversight onto the design of user instructions, account limits and system controls.
The AI Partition Creates a Separate Capital Boundary
Clients can fund the AI Partition by transferring money from their existing Interactive Brokers account, without opening a new account or making an external deposit. The partition is intended to set aside capital available to the automated process. Interactive Brokers’ educational materials say launch limits apply and can vary according to account size, eligibility and other restrictions. In the examples described by the firm, the amount assigned to the partition can be capped, including for larger accounts.
A separate partition can help a client limit how much capital an AI platform can act on, but it does not eliminate market risk. Orders can execute at unfavorable prices, a strategy can behave differently in unusual conditions and the value of an asset can move sharply before a system responds. A cap on allocated funds limits the scope of exposure; it does not ensure a profitable result or prevent losses within the amount allocated.
The firm says its safeguards are subject to account restrictions and that eligibility varies by Interactive Brokers affiliate, client country and regulatory requirements. It also says funding and account limits apply at launch. These qualifications mean the feature is not necessarily available to every client or for every instrument. The release refers to supported stock and equity-option orders; it does not establish that all products accessible through the brokerage can be traded autonomously.
Brokerage Controls Will Be Tested Alongside AI Capability
For brokers, the product raises questions about how instructions are translated into orders, how clients understand the scope of delegated authority and how errors are handled. Natural-language instructions are accessible, but their meaning can depend on details such as the price reference, time window, position size and response to incomplete data. A rule that sounds clear to a person may still require precise definitions before it can be executed consistently.
That is particularly relevant for options. A covered-call instruction depends on the underlying position, contract selection, expiry, strike, implied-volatility measure and order price. The broker says supported orders can be submitted when authorized criteria are met, but the client remains responsible for setting the criteria and understanding what the resulting trade may do to the portfolio. A seemingly simple trigger can produce a complex exposure.
Autonomous activity also creates a monitoring question. Clients need to know what actions the AI took, which rule triggered an order and whether the order was filled, rejected or modified. The release describes the dedicated partition and eligibility controls but provides limited detail on audit logs, error correction or procedures for disputed trades. These are operational details that matter when a system can act without a per-order human review.
Part of a Wider AI Integration Push
Interactive Brokers lists other AI tools alongside the new feature. Its AI integration supports connections with platforms such as ChatGPT, Claude, Grok and Perplexity through the Model Context Protocol, while AI Screeners let users describe securities they are looking for in natural language. Investment Themes and Connections link securities to industries, funds, derivatives and related products.
The new service sits within a wider market effort to connect conversational systems to trading accounts. FinanceFeeds has covered Interactive Brokers’ earlier additions of ChatGPT and Grok to its AI integrations, as well as other broker and platform experiments that let AI assistants prepare or submit trading instructions. The move from assistance to autonomous submission is a material step because it changes the point at which a human must act.
For clients, the appeal is the ability to monitor markets continuously and apply rules without manually repeating tasks across time zones. The trade-off is that a client must define boundaries that are robust enough to govern the system when conditions change. Interactive Brokers’ AI Partition offers one capital-control mechanism, but the quality of the instructions, the supported order types and the client’s understanding of the risks remain central to how the feature works in practice.
