You are a bit naive. They are making a ton of money with this dark pattern. As others have said Free-to-100K is not in the most generous realm of expectations. Its also why they have been doing the refunds as long as AWS has been a thing. They know it will not hold up in court. Not a month goes by without some HN story about something like this post.
They do this and make it easy to get a refund because for every demo account that does it some bigger account accidentally gets billed 10K and they have to pay for it. They have skin in the game and cannot risk their account being down for any time period.
I have had a “larger account” when I was at startup and was able to ask for a refund for a business.
As I asked before, if what is causing overages is not web requests but storage should they just delete everything?
Not sure what you define as larger account.
Counter real world example. I was doing some consult work for a place that had a 9k unexpected charge on AWS. They asked about it and I explained about how they could dispute it. They said ugh never-mind and just paid it. FYI it was a charity which I've since learned its common for charities to be wasteful like this with money since they are understaffed and OPM(Other peoples money)
So how is that a counter example? The client never asked for a credit. Since the startup I worked for, I have been working in AWS consulting - first directly at AWS (Professional Services) and now a third party consulting company.
While I have no love for AWS as an employer, I can tell you that everyone up and down the chain makes damn sure that customers don’t waste money on AWS. We were always incentivized to find the lowest cost option that met their needs. AWS definitely doesn’t want customers to have a bad experience or to get the reputation that it’s hard to get a refund when you make a mistake.
Of course AWS wanted more workloads on their system.
Semi related, AWS will gladly throw credits at customers to pay for the cost of migrations and for both inside and outside professional services firms to do proof of concepts.
You, but shorter: It can't be done perfectly in 100.0% of all possible circumstances, so better to do absolutely nothing at all. On an unrelated note, this strongly aligns with their economic interests.
For storage specifically, in that circumstance, if you weren't hellbent on claiming otherwise: it's easy to figure out what to do. For storage: block writes and reach out to the customer. Also, people are extremely unlikely to accidentally upload eg 250tb which is how you'd get to, say, $200/day. Whereas similar bills are extremely easy to accidentally create with other services.
It's totally reasonable to want spend limits firmer than AWS' discretion, which they can revoke at any point in time for any reason.
Just introduced in July. It won’t let you go over $200
https://aws.amazon.com/free/