Elon Musk made a bold statement: if the AI agent Grok, while managing a user's bank account, makes a mistake and loses funds, the company xAI will reimburse the losses. However, a careful look at the actual terms of service makes one doubt the generosity of this promise.
At the center of the discussion was an experiment conducted by an investor under the nickname Teslaconomics. He asked on social network X whether anyone had connected Grok to their bank account. According to him, such an agent can track expenses, pay bills, and identify suspicious charges. Moreover, the user even wondered whether he still needed a personal banker. However, his colleague expressed reasonable concerns that AI access to finances could lead to unpredictable consequences.
Musk was not bothered by such doubts. He publicly promised to compensate for losses if Grok makes a mistake. But let's look at the facts. The beta version of the product was released on August 11. In xAI, they explained that the agent works around the clock on its own cloud server, simulating the actions of an ordinary user, even while the owner sleeps. This sounds impressive, but this is exactly where the main catch lies.
Hidden Limitations: $100 Liability
Musk's promise looks like insurance, but the company's legal documents say otherwise. In the xAI user agreement, it is explicitly stated that the results and actions of agents are provided "as is." The maximum claim amount is limited to the amount of paid fees or $100—whichever is greater. Given that access to the bot costs $30 per month under the SuperGrok plan, that adds up to $360 per year. But this is only a small part of the potential losses that could arise when managing an account.
Musk's response on X does not change these terms. Until xAI formalizes the guarantees in writing, any compensation remains at Musk's personal discretion. The risks are also compounded by US banking rules. Regulation E protects customers from unauthorized charges, but if the owner himself gave the bot access to the account, such transactions are no longer considered fraudulent.
Real Threats: Prompt Injections and Vulnerabilities
Skeptics rightly point to recent incidents. In May, a malicious NFT "hid" instructions that forced the AI to transfer money—this is a classic example of a prompt injection. As a result of the attack on the Bankr wallet associated with Grok, about $150,000 was withdrawn. Later, approximately 80% of the stolen funds were recovered, but the bad taste remained. A few weeks later, 14 more wallets were affected, and Bankr promised full reimbursement. However, there is not a single proven case of Grok making a mistake with a real bank account yet.
The experiment with the bot is not only a test of technology but also part of xAI's marketing strategy, which seeks to integrate the product into everyday financial management. The first real mistake will show the true value of Elon Musk's words. The company can either silently refund all the money or publicly acknowledge the $100 liability cap. Investors and users should be extremely cautious when trusting AI with access to their funds.
My professional opinion: such statements are a classic PR move designed to create an image of innovation. The real legal protection for users is minimal, and the risks associated with the security of AI agents remain extremely high. Until xAI offers transparent and enforceable guarantees, using such bots for financial management is an unjustified risk.