Amazon is fixing a bug in its AWS billing portal that showed some customers wildly wrong cost estimates, with figures ranging from a few million dollars to nearly $2.5 billion for a single month. TechCrunch reported the issue on July 17, 2026, after affected customers started posting screenshots of their inflated estimates on Reddit.

Amazon confirmed the problem on its AWS status page on Friday, July 17, saying it began seeing inaccurate billing data as of late Thursday, July 16. The company said the bug is in the part of its system that calculates cost estimates, and that rolling back a recent change did not fix the issue as of Friday morning. Amazon expected the problem to last several more hours.

The estimates, while alarming, carry no actual charge. Amazon stated that the figures "do not reflect actual usage and charges." In other words, the system that tracks how much customers really used is working fine. The problem is in the layer that turns that usage data into dollar amounts shown on the billing screen. Think of it like a cash register that shows the wrong total while the scanner correctly records each item you bought: the receipt would be wrong, but the store knows what you actually purchased.

The scale of the distortion varied widely. One AWS customer posted a Reddit screenshot showing a monthly billing estimate of close to $2.5 billion. Others reported estimates in the range of a few million to hundreds of millions of dollars. The common thread across affected accounts is that the displayed figures bore no plausible relationship to what the customers actually used.

TechCrunch's reporting drew primarily on Amazon's own AWS status page updates and customer posts on Reddit. An Amazon spokesperson did not immediately return TechCrunch's request for comment. The AWS Health Dashboard had separately reported an operational issue affecting the us-east-1 region on July 16, though the available information does not confirm a direct link between that issue and the billing bug.

For anyone who has managed cloud budgets, the distinction between a tracking failure and a calculation failure matters. If the raw usage data is intact, customers will not be overcharged on their actual invoices. The estimates in the portal are informational, not the basis for payment. The damage in this scenario is mostly operational: finance teams and automated alert systems that check the billing data can trigger on the corrupted estimates, generating false alarms across an organization.

The broader context here is that cloud cost management has become a serious discipline of its own. Companies spending millions on AWS rely on billing data to make decisions about scaling, cost optimization, and budget limits. Many teams build automated alerts and dashboards that treat the billing estimates as near-real-time truth. When those estimates can swing to billions of dollars without any change in actual usage, the reliability of the entire system built on top of them comes into question. A billing estimate that shows $2.5 billion where the real figure is presumably a small fraction of that is not a rounding error; it is a total failure of the calculation path, and the fact that rolling back the change did not immediately fix it suggests a meaningful fault in how the system handles updates.

Amazon's prompt public acknowledgment and its statement that actual charges are unaffected are the right steps. The status-page transparency, however, follows AWS's standard incident-communication pattern: periodic updates with limited technical detail on what went wrong. For a billing-system failure of this magnitude, affected customers will reasonably want a follow-up explanation of how the system produced such enormous errors and what changes will prevent it from happening again.

The episode is, at its core, a contained incident with no financial impact on customers. It is also a reminder that the billing layer of a major cloud provider is itself a complex system that can break, and that anyone relying on its data should treat any single source of cost information with a degree of skepticism.