Best Chat-Based Investing Apps in 2026

A chat-based investing app is an app whose main interface is a conversation. But there is a real gap inside that category between apps you talk to and apps you invest through. Only some let a plain-language request turn into an order:

• Chat-first, act on it (Magnifi). Built around chat for markets: you describe what you want, it helps you screen and discover funds, ETFs, and stocks, and its in-app investing features let you buy what you find. The conversation leads into the trade.
• Assistant inside a brokerage (Public, via Alpha). A full investing app where you hold a real account, with an AI research assistant layered in. You ask Alpha, then place the trade in the same app.
• General assistant, no brokerage (ChatGPT). A powerful explainer people already use to reason about investing, but with no account view and no ability to place an order on its own.
• Chat on the broker you already own (Walnut). A chat that connects the real brokerage you already own, turns the conversation into a thematic portfolio, and prepares the trades you approve, so you act on an outside account rather than moving money into a new one.

An app you only talk to is useful for learning and planning. An app you invest through closes the loop between the question and the position. Both matter; they just do different jobs.

Under the hood, “invest through chat” is a few steps stitched together so they feel like one conversation. Knowing the steps makes it obvious why some apps can close the loop and others stop halfway:

• You state intent in plain language. “Find me a low-cost clean-energy ETF” or “how exposed am I to AI infrastructure?” The app parses that into a research task instead of making you learn a screener.
• The AI researches and proposes. It screens securities, summarizes context, and proposes something concrete: a fund, a set of tickers, or target weights. A good one shows its reasoning rather than a bare output.
• An account gets involved. This is the fork. Some apps hold your money in-house (Public), some connect a broker you already own through a regulated aggregator (Walnut), and some have no account at all (ChatGPT), which is exactly why it cannot place the trade.
• You confirm, and the order routes. The responsible pattern keeps you in the loop: the app prepares the order and you approve it, rather than trading silently. Walnut is read-only by default and requires your approval for every trade.

The step that separates “chat that invests” from “chat that only talks” is the third one: whether there is a real account wired in, and whose account it is.

These are the two apps here where a conversation is meant to end in a purchase, each pointed at a different account model. Magnifi is chat-first for markets and fund discovery, with in-app investing to act on what you find. Public pairs a full brokerage that holds your money with Alpha, its AI research assistant, so you ask then trade in one place.

These are the right call when your question matches the app: “which ETF fits this exposure” for Magnifi, or “research and trade in one brokerage” for Public. They are the wrong call when you want to keep trading through the specific outside broker you already use, with the conversation grounded in those existing holdings.

That last gap is where Walnut sits next to Magnifi and Public. Like Magnifi, Walnut is chat-first and you ask in plain language; like Public’s Alpha, it puts research and action close together. The difference is whose account you act on: where Magnifi and Public route into their own investing, Walnut connects the outside broker you already own, reads it read-only by default, frames each holding against the S&P 500, and turns the chat into a thematic portfolio whose trades you approve. Walnut is not an investment adviser.

For many people, the first chat-based way they think about investing is just ChatGPT, and it is genuinely strong for explaining concepts and reasoning through a decision. The catch is the one that defines this whole page: on its own it cannot see your brokerage or live prices and cannot place an order, and it can state wrong figures confidently, so verify anything specific.

The practical takeaway: use it to learn and to think, not as a source of truth on your actual numbers, and not as a way to execute. If you want one of these models to reason over your real holdings and then help you act, you need an app that connects your accounts to it. Walnut is one such app: it pipes your connected brokerage into a chat through Claude or ChatGPT (or a built-in assistant), so the same models that explain a concept in the abstract can reason over the positions you actually own, each framed against the S&P 500, and then prepare the trade for your approval.

To be upfront, since this is our site: Walnut is the AI investing assistant that talks to the broker you already have and places the trades you approve. It has already appeared next to Magnifi, Public, and ChatGPT above because it overlaps with each: it is chat-first like Magnifi, pairs research with action like Public’s Alpha, and talks through Claude or ChatGPT like the general assistant. What makes it its own category is the combination: you invest through the chat, but on the broker you already own. It leads in that narrow lane rather than overall.

The distinctive part is that the chat knows your real positions, frames each one against the S&P 500, and can become a thematic portfolio you act on at your own broker. It is read-only by default, every trade needs your approval, and Walnut is not an investment adviser. But it is the wrong choice in several cases, and another app on this page fits better:

• You want to be hands-off. Walnut keeps you in the loop on every trade, so it adds work you may not want. A traditional robo-advisor is built precisely so you do not have to be involved; see the AI robo-advisor alternatives if you want a middle ground.
• You do not already have a brokerage. Walnut sits on top of an existing account, so it needs one. Public, which holds your money in-house, suits better if you want to open and fund a new account.
• You mainly want to discover funds and ETFs. Magnifi is built around that lane; Walnut narrows to the broker you already own rather than broad fund screening.
• You only want to explain a concept or reason through a scenario. ChatGPT on its own is free and excellent for that; you do not need a connected app until you want to act on your real holdings.

Walnut is also not an all-in-one bank: it leans on web and price data rather than a proprietary filings corpus, and because broker feeds rarely pass cost basis it frames returns as window returns rather than realized profit and loss, and says so.

The fastest way to choose is to name what you are trying to do, then pick the app built for that. There is no overall number one; Walnut leads only in its own lane (investing through chat on the broker you already own), not across the board.

• You want to discover funds and ETFs in chat, then buy them. Magnifi is built around conversation for that lane, with in-app investing to act on it.
• You want AI research and trading inside one brokerage. Public pairs its Alpha assistant with a full in-house account so you ask then trade in one place.
• You want to learn a concept or reason through a decision. ChatGPT is the general assistant most people already use; verify any specific figures, and remember it cannot place the trade itself.
• You want to chat on the real broker you already own and act on it. Walnut connects your brokerage, lets you research through Claude or ChatGPT framed against the S&P 500, and prepares the thematic-portfolio trades you approve.

Once you know whether you want to discover, hold in-house, or act on your own broker, a few practical filters narrow it the rest of the way:

• Can you actually invest through the chat? Some apps stop at research; some close the loop into an order. If you want to act, rule out the ones that only explain.
• Whose account do you act on? In-house (Public), a broker you already own (Walnut), or none at all (ChatGPT). If you want to keep your existing broker, that points to a connected app.
• How does account access work? If an app touches your money, prefer regulated aggregation, read-only-by-default access, and explicit approval for any order. Walnut keeps access read-only by default and approves every trade with you.
• Hands-off or hands-on? A robo-advisor runs on autopilot; a chat-first app keeps you in the loop. Pick based on how much control you want over each decision.
• Cost model. Free tier, flat subscription, or a percentage of assets. ChatGPT and Walnut have free tiers; verify current limits before relying on them.
• Does it stay descriptive? A trustworthy investing app explains and frames trade-offs without pretending to be your adviser. Be wary of anything promising guaranteed market-beating returns.

There is no single best chat-based investing app, because they close the loop in different ways. Magnifi leads for discovering and buying funds and ETFs by describing what you want, Public (Alpha) for keeping AI research and trading inside one brokerage, and ChatGPT for general explaining and reasoning, though it cannot place the trade itself. Walnut is the one that lets you invest through the chat on the broker you already own: it connects your brokerage, lets you talk through Claude or ChatGPT, frames each position against the S&P 500, and turns research into a portfolio you act on, while keeping you in the loop on every trade. Pick by whether you want to discover, hold in-house, or act on your own broker. Walnut is not an investment adviser.

For the wider field, see the best AI investing apps roundup, or, for the specific case of managing a broker you already own, our guide to the best AI investing app by use-case.