Elon Musk pledges compensation for Grok Bot financial losses amid liability concerns
Elon Musk promised to reimburse users if his Grok Bot AI agent loses money while managing their bank accounts, but xAI's terms cap liability at $100 or fees paid, whichever is greater. Security experts noted that prompt injection attacks have previously targeted Grok-connected wallets and could pose risks to bank account access.
What happened
Grok Bot launched in beta on August 11 and can be connected to user bank accounts to track spending, pay bills, and catch unusual charges. Elon Musk pledged on August 27 that xAI would reimburse users if the AI agent loses money while managing their accounts.
xAI's consumer terms, however, cap liability at the greater of fees paid or $100, meaning a year of the $30 monthly SuperGrok subscription totals only $360 in coverage. A reply on X (Twitter) does not rewrite the contract, and any refund rests on Musk's goodwill unless the guarantee is put in writing.
Reported impact
No verified case of Grok Bot mishandling a real bank account has surfaced. In May, a prompt injection attack through a malicious NFT drained roughly $150,000 from a Grok-linked Bankr wallet, with about 80% of funds later recovered.
Index analysis
Prompt injection, a technique using hidden instructions to trick an AI into moving money, has previously targeted Grok-connected wallets. When users hand over bank account login credentials to a bot, standard fraud protections under Regulation E, which covers unauthorized electronic transfers, may not apply.
Governance controls
xAI's consumer terms offer outputs and agentic actions on an as-is basis with a liability cap. US federal banking rules including Regulation E provide protections for unauthorized payments, though these protections may not extend to transfers made with customer-granted access.
Classification
Sources and evidence
This record was researched and written by the Index, based on the following original reporting.