I started thinking about this question relatively recently. Competitive programming platforms (like Codeforces, LeetCode, HackerRank, and so on) don’t provide open tests. Mostly they give us a few test cases so we can run our code against them, get a rough idea that it works, and then submit. But this submission will be run through hundreds or thousands of additional tests that the platform keeps private. Sometimes on certain platforms we can see which test failed, but that doesn’t change the essence: all these tests are company property. Recently, while reading through an old chat history, I came across a message written several years ago.

who said such bullshit? they’re closed. it’s their strategic commodity

– Unknown user

And with each passing day I’m starting to believe this more and more. If you really think about it, we can come up with tons of reasons why they’re kept private. Moreover, they might even have more value than the problems themselves. An error in a solution, a specific failed test case, a pattern of attempts - all of this can say a lot about a person’s thinking style. Furthermore, you could plug AI into this task and sell information about all your mistakes and submission attempts to interested parties.

Or is this already happening? More likely yes than no, but what I know for sure is that no matter how open a competitive programming platform might seem, in reality they’re all heavily centralized and generally closed.

btw, I want to at least try to take a step in the opposite direction and attempt to do something towards a possible solution to this problem.