Your fridge could be the next crypto trader—CFTC asks who’s liable when AI goes rogue 🤖

Commodity Futures Trading Commission Chair Michael Selig said Wednesday that autonomous AI agents could eventually use staked Ether to buy groceries or launch memecoins, describing the scenario as “no longer science fiction.” Selig made the comments during the CFTC’s inaugural Frontier Forum in Washington, DC, as regulators consider how existing rules apply to software that can hold assets and execute trades without direct human intervention.

“We’re gathered here today to discuss a technology that’s rapidly transforming the financial markets. President Trump calls it ‘super intelligence,’ or SI, because ‘it’s not artificial.’ It’s as real as the printing press, the steam engine, and the internet,” Selig said. He said CFTC rules were built around “identifiable market participants” performing specific financial functions, a structure challenged by autonomous software.

Selig outlined unresolved questions about developer responsibility, liability and regulatory oversight. “Consider an SI agent that identifies trading opportunities and executes transactions on behalf of a user,” he said. “Does the developer of that agent become subject to registration? Who is liable if the agent goes rogue? How can regulators know who the agent, or suite of agents, is trading on behalf of? Do we need a new flavor of know-your-customer, maybe KYA or know-your-agent?”

An eToro survey published in September found that 79% of US millennial investors and 77% of Gen Z investors already used or were open to using AI to select or adjust their investments. Selig said AI could give smaller traders and businesses access to risk-management tools once reserved for larger institutions, while also creating risks from unexpected reactions to markets, correlated behavior among users of similar models and faster transaction speeds that could complicate surveillance.

Animoca Brands Chairman Yat Siu recently raised a similar concern, telling Cointelegraph that a thousand people asking AI about the same trade could receive roughly the same answer, while users’ own knowledge and follow-up questions help shape more personalized strategies. Last year, the Bank of England’s Financial Policy Committee also warned that widespread use of similar models could amplify market shocks.

Crypto exchanges are already developing tools for autonomous trading, with Coinbase introducing Agentic Wallets in February. Selig said regulators must adapt to a financial system in which AI agents can act autonomously, raising questions over registration, liability and the identification of the parties behind agent-driven transactions.