Disciplined AI agents are the disruptor needed to break the exchange churn model

Opinion Share Share this article Copy link X icon X (Twitter) LinkedIn Facebook Email Disciplined AI agents are the disruptor needed to break the exchange churn model Programmable incentives that allow independent trading agents to earn only when portfolios rise will create a fairer market for retail customers, explains Naja. By Saad Naja | Edited by Betsy Farber Updated May 28, 2026, 2:29 p.m. Published May 28, 2026, 2:13 p.m. 4 min read Make preferred on (Pakin Songmor/Getty Images) All within a matter of weeks, Anthropic unveiled new agents for finance, Circle launched nanopayments, MoonPay launched a debit card for agents and Gemini launched agentic trading, signaling the agentic finance fight is here. Whilst the products are new, the underlying business model remains the same. Every exchange and brokerage earns more when customers trade more, and the data on what that does for customer portfolios is unambiguous. Ultimately, agentic rails have arrived faster than incentives have changed. The perverse incentives exchanges hope you miss The conflict is structural to the industry. Brokerages and exchanges don't need customers to win, they need them to keep trading. Crypto exchanges and neobrokers made trading faster, cheaper and frankly, more addictive. The commercial reality is that banks profit when you stay, exchanges profit when you trade, and AI models profit when you prompt. The agent you can trust with your hard-earned capital sits outside all three. An independent agent paid only when the customer’s portfolio wins threatens the current incentive structure of brokerages and exchanges. The truth is, zero-commission trading isn't free. In 2025, U.S. market makers paid more than $4.9 billion for order flow in U.S. equity and options, up from approximately $3.8 billion in 2021 across the 12 largest U.S. brokerages. The same principle applies to crypto. The derivatives volume from Q1 of 2026 reached about $18.6 trillion , 70% of global crypto trading, with perpetuals dominating spot trading. Exchange economics reward trading velocity over disciplined decision-making. At peak, Robinhood relied on more than 75 percent of its revenue from payment for order flow (PFOF), the hidden backbone of "free" trading, in which market makers pay brokers to route customer orders. Every broker using this incentive model needs customers to trade often, even though frequent trading works against long-term returns. Advisory isn’t better. Robo-advisors charge 0.25 percent of assets a year, whether the account is up or down. Human advisors charge around 1 percent , billed against the principal even in down years. The extraction is built into the model by design: the advisor gets paid even when the customer loses. Less exchange friction makes bad trades easier to repeat The harsh truth is that exchanges need customers to trade more, not win. When retail investors lose, the exchanges still get paid. PiP World research found 74% to 89% of retail users lose money trading. Platforms charge at every step, and an AI-enabled exchange could just route you back to the same losing trade faster. The April 14 SEC approval of FINRA's elimination of the Pattern Day Trader rule removed the $25,000 minimum-equity friction. Removing the friction results in more trades, which creates more order flow. More order flow means more money for the broker, whether the customer’s profit and loss (P&L) is up or down. Enter AI agents, paid to improve customer P&Ls The disruptor to this vicious cycle for retail traders is the agent built to do what the existing exchange model avoids: trade less, size down, wait and protect customers from their worst impulses. In volatile markets, the best move is often refusing the bad trade, cutting exposure before emotion takes over. Ultimately, holding discipline when the market wants a reaction. Discipline is hard to sell for an exchange because it shrinks order flow. An agent that earns by protecting customer P&Ls breaks the current incentive model. The next battleground is who profits from the agents' order flow Regulators are squeezing the old "free trading" model. The EU's PFOF ban takes effect June 30, 2026, removing the revenue line behind "free" trades for German and Austrian neobrokers. Trade Republic, a European savings platform, has already found another route to secure a BaFin license to internalize order flow. Whilst TradFi scrambles to patch the leaks, crypto builders are racing to rebuild onchain rails for AI agents. In markets with tiny spreads, fragmented liquidity and millisecond execution, agents transact via nanopayment infrastructure like Circle's protocol. Gas-free trading on perpetual DEX Hyperliquid cuts friction, but maker-taker fees still apply. The real fight ahead isn't who removes friction, but who profits when agents start hammering these frictionless rails with high-frequency trading. Independent programmable agents are better middlemen The exchanges and brokers have spent years making money from customers trading more, understanding less and absorbing tiny costs they barely notice. Every agent built by an exchange will inherit the exchange’s incentives. Would an exchange build an agent that sends trades through a cheaper competitor’s rails? Not voluntarily. Whereas an independent agent has one job: grow and protect the customer’s portfolio, routing trades where they work hardest for the customer. Programmable incentives encoded into smart contracts tie the agent’s incentives to portfolio gains. The customer can see where the money goes, verify what the agent gets paid, when and why. With independent agents, the customer keeps more of the value that used to leak to the exchange through order flow, spread markups and idle-cash interest onto the exchange. The agent is rewarded for disciplined trading, not constant trading. It can trade often when the signal is strong, cut exposure when risk rises and sit out when the market is just noise. The first agentic platform that proves this alignment onchain will give retail investors a fairer counterparty, whose economics finally move in the same direction as theirs. Artificial Intelligence Note: The views expressed in this column are those of the author and do not necessarily reflect those of CoinDesk, Inc. or its owners and affiliates . More For You A bipartisan bridge to the future: Why the Senate must finish the job on digital Assets By Ji Hun Kim | Edited by Betsy Farber May 27, 2026 The Clarity Act’s recent markup proved that the momentum for regulation is there. It is imperative that Congress move the bill forward to establish rules this generation needs and a framework the next will inherit, urges Kim. Read full story Latest Crypto News Gemini taps SpaceXAI to build a personalized prediction markets feed 33 minutes ago 'Debasement trade’ falls out of favor as inflation fears cool, JPMorgan says 55 minutes ago Asset manager Grayscale delays IPO plans as crypto listing boom loses steam 57 minutes ago VanEck launches first U.S. spot BNB ETF on Nasdaq 57 minutes ago Bit Digital bought $20 million ETH for first time since October before 15% plunge 1 hour ago What's next as hot money cycle has gone from crypto to gold to AI to memory 1 hour ago Top Stories Bitcoin drops below $73,000 as U.S. strikes on Iran spark $1 billion liquidations 10 hours ago White House reviews CFTC prediction-market rule as Trump backs federal control 5 hours ago Samsung units to buy $408 million stake in South Korea’s biggest crypto exchange 3 hours ago Diverging trends: Ether slides below $2,000 while futures open interest hits record high of 16 million ETH 7 hours ago Standard Chartered backs $4,000 ether as retail piles into the sub-$2,000 drop 3 hours ago The crypto industry’s massive political war chest is starting to lean Republican ahead of midterms 17 hours ago