Anthropic confirmed a massive billing error after its system tried to charge a South Korean free-tier user $16.6 million, despite zero API usage and no registered card. The first invoice arrived on July 7, totaling $1,669,875, and a second one multiplied the figure almost tenfold to $16,627,739 less than 24 hours later, with no activity recorded on the account. The developer initially suspected phishing, but the emails came from Anthropic's official domain and used legitimate Stripe infrastructure; his dashboard showed zero API usage, no billable keys, and no payment method on file. His bank declined two overseas charge attempts on July 8 because the amounts exceeded the card's per-transaction limits, but the repeated attempts still resulted in his primary credit card being blocked. Anthropic confirmed the mistake on July 12 and assured no money was collected, attributing the invoices to an incorrect automatic credit reload setting, which it disabled as a precaution, and clarified the incident did not involve unauthorized access. The user spent four days and around 18 emails seeking written confirmation that the invoices were void. Separately, audit startup Vaudit reviewed $34 million in AI invoices from 60 enterprise clients and found roughly $1.7 million in overcharges, representing an error rate of about 5%, with affected clients including Panasonic, HP, and Honda.