Elon Musk is once again shocking the public, this time by promising to compensate users for losses if his AI agent Grok, which manages a bank account, makes a mistake. It sounds bold, but upon closer inspection, behind this loud statement lies a whole range of legal and practical nuances that cast doubt on the unconditional nature of such guarantees.
Experiment with a bank account: boldness or recklessness?
One investor, known by the nickname Teslaconomics, decided to test the capabilities of the new product. The question was simple: had anyone already connected Grok to their bank account? As conceived, the agent should track expenses, pay bills, and even identify suspicious transactions. Teslaconomics himself wondered whether it was time to give up the services of a personal banker, but his colleague expressed reasonable concerns that an AI with access to finances could cause havoc.
Musk was not fazed by such doubts. He confidently stated that xAI would reimburse losses if Grok made a mistake. The beta version of the product was already released on August 11. The company explained that the agent works around the clock on its own cloud server, visits websites like a regular user, and continues to perform tasks while the owner sleeps. This is an ambitious attempt to integrate AI into everyday financial management, which fits perfectly into Musk's global strategy, including the launch of the X Money service with direct transfers in June.
Limitation of liability: $100 — and that's the whole guarantee
Musk's promise looks like insurance, but the company's documentation says otherwise. The xAI user agreement clearly states: the results and actions of agents are provided "as is." Moreover, the maximum claim amount is limited to the amount of fees paid or $100 — whichever is greater. Given that access to the bot costs $30 per month (or $360 per year under the SuperGrok plan), the compensation amount looks more like a mockery than real protection for those who entrust their savings to AI.
Musk's response on X does not change these terms. Until xAI formalizes the guarantee in writing, any compensation remains at Musk's discretion.
Risks and real threats
The situation is further complicated by US banking regulations. Regulation E, which protects customers from unauthorized debits, may not apply if the owner himself granted the bot access to the account. Such a debit would no longer be considered unauthorized. Skeptics also point to recent incidents: in May, a malicious NFT "hid" instructions that forced AI to transfer money — a classic example of prompt injection. As a result of an attack on the Bankr wallet associated with Grok, about $150,000 was withdrawn, although approximately 80% of the funds were later recovered. A few weeks later, 14 more wallets were affected, and Bankr promised full reimbursement. So far, there is not a single proven case of Grok making a mistake with a real bank account, but the damage could be far more serious than with a crypto wallet.
The Grok experiment is partly also advertising: xAI is clearly striving to integrate the product into everyday financial life. But the first real mistake will show what Elon Musk's words are worth. The company will either silently return all the money or publicly limit itself to that very $100 cap.
My expert opinion: For now, this is nothing more than a marketing ploy. xAI's legal framework leaves a huge "gray area" for refusing compensation. Trusting an AI agent to manage your account until the guarantees are documented and cover real amounts means playing Russian roulette with your finances. Keep an eye on developments, but be extremely cautious.